ArticlePDF Available

Nutritional Status of Papua New Guinea's Population and Its Determinants

Authors:

Abstract

Data from a nationally representative household survey in 1996 are used to describe the nutritional status of the rural and urban popu lations. The indicators examined are the per capita availabilities of calories and protein, the energy density of the diet, the standardised height of young children and the body-mass index of adults. Multivariate analysis shows that nutrient availability rises by between four and seven percent for every ten percent increase in household economic resources, suggesting that economic growth can have beneficial effects on nutrition. The response of nutrients to increased household resources is highest in the r ural sector and is higher for protein than for calories. The hypothesis that rural households oriented towards tree crop production have lower nutrient availability than do households oriented towards food crop production is not strongly supported. In cont rast to the direct effects of (especially mother's) education on children's heights, educational effects on nutrient availability work mainly through raising household incomes. A basic constraint on raising household incomes, and on improving health and nutrition status, is lack of access to public services.
Nutritional Status of Papua New Guinea’s Population and Its Determinants
Paper Presented at the
Papua New Guinea Food and Nutrition 2000 Conference
26-30 June 2000
Rose Kekedo Convention Centre
PNG University of Technology
John Gibson
1
University of Waikato
Abstract
Data from a nationally representative household survey in 1996 are used to describe the
nutritional status of the rural and urban populations. The indicators examined are the per capita
availabilities of calories and protein, the energy density of the diet, the standardised height of
young children and the body-mass index of adults. Multivariate analysis shows that nutrient
availability rises by between four and seven percent for every ten percent increase in household
economic resources, suggesting that economic growth can have beneficial effects on nutrition.
The response of nutrients to increased household resources is highest in the rural sector and is
higher for protein than for calories. The hypothesis that rural households oriented towards tree
crop production have lower nutrient availability than do households oriented towards food crop
production is not strongly supported. In contrast to the direct effects of (especially mother’s)
education on children’s heights, educational effects on nutrient availability work mainly through
raising household incomes. A basic constraint on raising household incomes, and on improving
health and nutrition status, is lack of access to public services.
Acknowledgements:
The data used in this paper were originally collected as part of a World Bank poverty assessment for Papua
New Guinea, for which financial support from the governments of Australia (TF-032753), Japan
(TF-029460), and New Zealand (TF-033936) is gratefully acknowledged. All views in this paper are those
of the author and should not be attributed to the World Bank.
1
Department of Economics, University of Waikato, Private Bag 3105, Hamilton, New Zealand.
Phone: (64-7) 856-2889. Fax: (64-7) 838-4331. E-mail: jkgibson@waikato.ac.nz.
1
Introduction
Researchers, planners and policy-makers in developing countries have long been concerned with
the issues of food security and malnutrition. Papua New Guinea also appears to be moving toward
viewing food security, rather than just food self-sufficiency, as the goal of national food policy.
This is a step in the right direction because food security the ability to command sufficient food
at all times is a more sensible goal than self-sufficiency. Command over food can be gained
using the most efficient and least risky means, whether that be through directly producing food or
else by producing cash crops to exchange for food. But food security, particularly when it is
reduced to a quantifiable measure such as whether calories available to a household are sufficient
to meet requirements (Garrett and Ruel, 1999), is only a means to an end. The overall goal should
be improved nutrition, and calorie availability is one of only several inputs into nutrition. Thus, a
preoccupation with food security and calorie availability may lead analysts to ignore other
determinants of nutritional status.
This paper uses data from a nationally representative household survey of Papua New Guinea in
1996 to describe differences in nutritional inputs and outcomes between the rural and urban
sectors. The nutritional outcomes considered are the height and stunting rates of young children
and the body-mass of adults. These outcomes are of considerable welfare significance, with, for
example, the consequences of stunting include increased risk of sickness and death (Chen,
et. al.
,
1980) and poor mental development (Grantham-McGregor,
et. al.
, 1996). The inputs to nutrition
that are considered are calorie availability, protein availability, energy density of the diet (i.e.,
calories per gram) and access to health services.
The results show that neither calorie availability, nor food security indicators based on calorie
availability, differ between urban and rural sectors. But there are substantial differences in
nutritional outcomes between sectors, and also in the non-calorie inputs to nutrition. These
significant urban-rural differences highlight the danger of concentrating exclusively on calorie-
based measures of food security. The results also show that nutrient availability responds strongly
to increased household incomes but a basic constraint on raising household incomes is lack of
access to services. Thus, improvements in infrastructure are likely to raise nutrient availability and
will also make non-calorie inputs to nutrition (e.g., health services) more readily available, which
should further improve nutrition.
The Papua New Guinea Household Survey
Data come from the 1996 Papua New Guinea Household Survey (PNGHS), a nation-wide
consumption survey conducted as part of a World Bank poverty assessment. The survey covered
a random sample of 1200 households, residing in 73 rural and 47 urban Census Units, selected
from the 1990 Census sample frame, stratifying by sector (urban and rural), by environmental
conditions (elevation and rainfall), and by the level of agricultural development.
2
Sampling
weights were generated from the variation between the Census estimates of the size of each
cluster and the actual size found in 1996, and from the deviation of the actual number of
households surveyed in each cluster from the target number. All results presented below take
account of the clustered, weighted and stratified nature of the sample.
2
Data from the survey and all survey documentation are freely available on the internet, at:
www.worldbank.org/lsms/country/png/pnghome.html.
2
The survey interviewed households at least twice, with the start of the consumption recall period
signalled by the first interview. The average length of time between interviews was almost two
weeks and the recall covered all food (36 categories) and other frequent expenses (20 categories).
The reported expenditures include the imputed value of own-production, net gifts received, and
food stock changes, so they should be a comprehensive monetary measure of consumption. Food
quantities were based on conversions from volume measures (households were given empty sacks
with marked graduations for recording garden produce), and the Pacific Islands Food
Composition Database was used to compute the nutrient quantities from the food quantity data.
3
The remaining components of household consumption were picked up by an annual recall,
covering 31 categories of infrequent expenses. An inventory of durable assets was also used to
estimate the value of the flow of services from these assets, including rental services from owner-
occupied dwellings.
Anthropometric measurements (weight and height) were made on all children in the surveyed
households who were age five years and under, and also on the parents of these children. Both
children and their parents were weighed and measured twice, once during the first visit to the
household and again during the consumption recall interview. This duplication allowed the
average of the two measures to be used, which should reduce the effects of measurement error.
Documentary evidence on children’s age (i.e., date of birth) was requested from the parents (e.g.,
birth cards, health books) but in some cases the only evidence was parental recall. Visits were also
made to health centres and hospitals to check birth records, and to churches to check baptismal
records.
Results The Calorie Puzzle
Average calorie availability appears to be similar in urban and rural sectors of PNG, at around
2660 calories per person per day (Table 1). Moreover, these averages don’t appear to disguise a
situation where one sector has a higher share of the population at the extremes of the distribution
(i.e. poorly-fed or overly-fed) because the degree of inequality in calorie availability in each sector
is very similar. In both the urban and rural sectors, approximately 42 percent of the population are
not meeting food energy requirements of 2000 calories per person per day.
4
Hence, a calorie-
based view of food security, such as that used by Garrett and Ruel (1999), would view urban and
rural PNG as equally deserving of attention.
3
One item where food quantities were not available was cooked meals eaten out of the home; calories from this
source were derived as the average “price” each household paid for all other calories plus a 50 percent premium to
reflect processing margins. A similar processing margin is assumed by Subramanian and Deaton (1996).
4
This target is equivalent to the nutritional requirement of 2200 calories per adult equivalent used by the poverty
lines in Papua New Guinea (Gibson and Rozelle, 1998) but is lower than some of the recommended daily
allowances published by the Department of Health.
3
Table 1.
Calorie availability in rural and urban sectors of Papua New Guinea, 1996
Rural
Urban
Papua New
Guinea
per capita daily calorie availability 2665
(76)
2645
(234)
2662
(74)
Gini coefficient on calories
a
30.3
31.8 30.5
% of population with <2000 calories
per day available
41.9
(2.4)
42.6
(5.8)
42.0
(2.2)
Note: Results calculated from 1144 households surveyed in 1996 Papua New Guinea Household Survey and
weighted to reflect the number of people in the sampling frame. Standard errors in ( ) adjusted for clustering,
stratification and sampling weights.
a
The Gini coefficient is a measure of inequality that ranges from 0 (perfect equality) to 100 (complete inequality
where one person controls all the calories and everyone else has none).
Despite the similarity in average calorie availability and in the proportion of the rural and urban
population lacking access to sufficient calories, there are big differences in nutritional outcomes for the
rural and urban population. Almost one-half of rural children appear to be stunted (low height-for-age)
but only one-fifth of urban children are stunted (Table 2)
5
. This estimate of the prevalence of stunting
was obtained by comparing the heights of the surveyed children with international growth reference
curves.
6
This comparison also showed that the average rural child in PNG is only 92.5 percent of the
median height of similarly aged (and gendered) children in the reference population, while urban
children average 97.3 percent of the median.
Low height-for-age of children is commonly assumed to reflect malnutrition due to the
accumulated effect of extended periods of inadequate food intake and past episodes of infection
and sickness. Hence, children’s height for age is an indicator which measures certain aspects of
food security, but it gives much different results than does the simple calorie availability measure.
Unlike the results for calorie availability, the child nutritional outcomes show a clear need to
direct resources into the rural sector so as to improve nutritional status. Looking at child height as
a nutritional outcome also suggests that there must be significant differences between urban and
rural sectors in some of the non-calorie inputs into this production process, in order to account
for the different outcomes.
5
The risk of stunting appears to be the same for boys as for girls (t=0.63) so the results in Table 2 are not broken
down by gender.
6
Previous anthropometric studies in PNG (Heywood, et al., 1988) have suggested that children from the highlands
do not fit the international growth curves very well, tending to be shorter (indicating malnourishment) but heavier
(indicating good growth). However, even if attention is restricted to children from the lowlands, the stunting rate is
40 percent in the rural sector and 20 percent in the urban sector, so the conclusion that nutritional outcomes are
worse in the rural sector holds.
4
Table 2.
Nutritional outcomes in rural and urban sectors of Papua New Guinea, 1996
Rural
Urban
Papua New
Guinea
Children age 0-5 years
Height as percentage of median for
age and sex in reference population
92.5
(0.4)
97.3
(0.6)
93.2
(0.4)
Percent who are stunted
a
47.0
(3.3)
19.8
(2.3)
42.9
(3.0)
Adults
Mother’s Body Mass Index (BMI)
b
21.6
(0.3)
25.3
(1.0)
22.1
(0.3)
Father’s Body Mass Index (BMI)
b
22.1
(0.3)
25.4
(0.6)
22.5
(0.3)
% of mother’s with BMI < 18.5
13.5
(2.6)
6.2
(2.6)
12.4
(2.3)
% of father’s with BMI < 18.5
4.5
(1.9)
1.4
(0.9)
4.1
(1.6)
Note: Results calculated from 969 children, 544 mothers and 454 fathers who were measured during the 1996
Papua New Guinea Household Survey. Estimates weighted to reflect the number of people in the sampling frame.
Standard errors in ( ) adjusted for clustering, stratification and sampling weights.
a
Height is more than two standard deviations below the median height for that age and gender in the reference
population used by the National Center for Health Statistics.
b
BMI = Weight (kg) / [Height (m)]
2
.
The survey also provides some information about the nutritional status of the adult population,
although the sample is non-random because the parents of young children are less likely to elderly
(the body mass index of elderly women can be low due to maternal depletion syndrome). With
this caveat in mind, the bottom part of Table 2 contains estimates of the average body mass index
for men and women, and the proportion whose body mass index is below 18.5, which indicates
chronic energy deficiency (Shetty and James, 1994). The average body mass index of urban males
is approximately 15 percent higher than for rural males (the gap is 17 percent for females). A rural
female is twice as likely as an urban female to have a body mass index so low as to indicate
chronic energy deficiency (and women are three-times more likely than men to suffer this
problem). Hence, the nutritional outcomes for adults also show that problems are considerably
worse in the rural sector, despite the similarity of calorie availabilities across rural and urban
sectors.
Results Other Inputs into Nutrition
The quality of the diet, as reflected in protein content and energy density, is considerably higher in
the urban sector. Protein availability and energy density for urban residents are approximately
50 percent higher than for rural residents. The average rural diet provides approximately 1.3
calories per gram, due to the dominance of root crops (which have an energy density of
approximately one calorie per gram). In contrast, urban diets provide around two calories per
5
gram due to the much higher content of cereals, fats and oils, and meats. Although an energy-
dense diet may cause obesity problems for adults, it can be a considerable advantage for young
children who may not be able to ingest all of the calories available from a bulky diet. Thus, the
higher energy density and higher protein content of urban diets may partly explain why child
stunting is much less prevalent in urban areas, despite the similarity in calorie availability.
In addition to the quantity and quality of the diet, child height also reflects past episodes of infection
and sickness. It is highly likely that rural children suffer a greater burden of infection because they have
poorer access to primary health care facilities. The average rural person has to travel for over one hour
to the nearest primary health care facility, compared with urban residents who typically are only 15
minutes from the nearest health care facilities.
7
This matters to nutritional status because people who
are sick may not be able to obtain the full nutritional benefit from their diet, making the comparison of
calorie availability a misleading indicator of nutritional status.
Table 3.
Non-calorie nutritional inputs in rural and urban sectors of Papua New Guinea, 1996
Rural
Urban
Papua New
Guinea
per capita daily protein availability 46.3
(1.9)
67.3
(3.6)
49.5
(1.8)
Gini coefficient on protein
a
37.6 35.8 38.0
Energy density (calories per gram)
1.27
(0.04)
1.92
(0.06)
1.35
(0.04)
Travelling time to nearest aidpost
b
70
15 60
Note: Results calculated from 1144 households surveyed in 1996 Papua New Guinea Household Survey and
weighted to reflect the number of people in the sampling frame. Standard errors in ( ) adjusted for clustering,
stratification and sampling weights.
a
The Gini coefficient is a measure of inequality that ranges from 0 (perfect equality) to 100 (complete inequality
where one person controls all the protein and everyone else has none).
b
Estimated from community-level data, which refers to the time taken using the means of travel commonly used by
people in the community. Where a health centre is closer than an aidpost, the time to the health centre is used.
An additional type of public service that is relevant to nutrition but is not included in Table 3 is
education and the provision of literacy services. Previous analyses on the 1996 survey data
indicate that there is a strong effect of maternal education in reducing the risk of child stunting
(Gibson, 1999) and there is a large gap in education levels between urban and rural sectors (and
also between men and women). The effect of maternal education persists even when controlling
for household incomes, so it is likely to reflect improvements in the efficiency and productivity
with which households use their resources to achieve improvements in nutritional outcomes.
7
Although it is likely that the urban health care facility will also be better equipped and have a wider range of
medicines, the survey cannot inform on this point because no data were gathered on the quality of public services.
6
The Determinants of Nutrient Availability
What determines nutrient availability at the household level? The results in Table 1 and Table 3
suggest that whatever the determinants, they will differ between urban and rural sectors, given the
similarity of calorie availability but the substantial differences in protein availability. Moreover,
while calories are slightly less equally distributed in urban areas, protein is less equally available in
rural areas and the overall degree of inequality in protein availability is higher. This likely reflects
the fact that the major sources of protein are purchased and that access to cash income in the rural
sector is less equal than is access to land (which is needed for growing the main calorie sources).
To uncover some of the determinants of nutrient availability, multivariate analysis has been carried
out, with the per capita availability of calories and protein regressed on per capita total household
expenditure, household size and demographic composition, and controls for the age, education
and income sources of the household head. In addition to these variables there are likely to be a
number of locational factors, including prices and environmental conditions, that influence
nutrient availability. In the absence of detailed information on these factors, one strategy is to use
dummy variables for each cluster in the sample (thereby soaking up all inter-community variation)
to give a set of within-cluster results.
The results of the regression analysis are reported in Appendix Table 1, and only the main points
and implications are highlighted here. The most important finding is that nutrient availability rises
by between four and seven percent for every ten percent increase in household economic
resources, suggesting that economic growth can have beneficial effects on nutrition. The response
of nutrients to increased household resources is highest in the rural sector and is higher for protein
than for calories. The estimated nutritional response to extra income is also higher when intercept
dummies for each cluster are used (the ‘within cluster’ results), so excluded price and
environmental factors are unlikely to be a cause of the results. Although rural households oriented
towards tree crop production appear to have lower nutrient availability than do rural households
oriented towards food crop production, this effect disappears once the control variables for each
cluster are included. A likely reason is that households where the cash income of the head is
derived mainly from sales of food crops are likely to be in more accessible locations and the
general rise in living standards associated with accessibility will also tend to raise nutrient
availability. Once locational effects are controlled for, the source of income for the household
head does not have a significant effect on nutrient availability.
The results from additional experiments with the regressions models are not reported in the
Appendix tables and are simply summarised. One key finding is that in contrast to the direct
effects of (especially mother’s) education on children’s heights, educational effects on nutrient
availability work mainly through raising household incomes. Once per capita expenditure is
controlled for, the effect of either women’s education or household head’s education is to reduce
nutrient availability and the most plausible explanation is that more households with more
educated members are more likely to be engaged in sedentary occupations where nutrient
requirements are lower.
7
Conclusions
An over-emphasis on calorie-availability measures of food security is likely to prove misleading in
Papua New Guinea. There is no difference between the rural and urban sector in average calorie
availability or in the proportion of the population with inadequate levels of calories available (a
common statistical indicator of food insecurity). Yet there are large differences in nutritional
outcomes between these sectors, with the risk of child stunting and chronic energy deficiency for
mothers being twice as high in the rural sector. There are several non-calorie inputs to nutrition
which appear to contribute to these poorer nutritional outcomes in the rural sector, including the
lower average and more uneven availability of protein in the diet, the lower energy density of the
diet and the poorer access to primary health facilities. The importance of these non-calorie inputs
into nutrition suggests that it is important to consider food security as a means to the end of
improved nutrition rather than an end in itself.
The results also show that nutrient availability responds strongly to increased household incomes
but a basic constraint on raising household incomes is lack of access to services. Previous results
estimated from the survey suggest that per capita expenditure (as a measure of household
economic resources) falls by 10 percent for every one hour increase in travelling time to the
nearest road or transport facility (World Bank, 1999). Thus, improvements in infrastructure are
likely to raise nutrient availability and will also make non-calorie inputs to nutrition (e.g., health
services) more readily available, which should further improve nutrition.
8
Appendix Table 1.
Nutrient availability regressions, Papua New Guinea, 1996
Within Cluster
Rural Urban
Rural
Urban
β
|
t
|
β
|
t
|
β
|
t
|
β
|
t
|
Calories
ln PCE 0.411
(9.98)
0.322
(4.21)
0.545
(15.1)
0.430
(13.1)
ln household size -0.261
(5.97)
-0.257
(3.50)
-0.160
(3.53)
-0.308
(3.68)
rf15+ 0.032
(0.26)
0.356
(1.89)
0.087
(0.66)
0.309
(2.34)
rf714 0.005
(0.03)
-0.548
(1.35)
0.001
(0.01)
-0.277
(0.82)
rf06 0.008
(0.05)
0.249
(0.73)
-0.127
(0.88)
0.477
(1.69)
rm714 0.071
(0.45)
-0.870
(1.93)
0.127
(1.05)
-0.469
(0.94)
rm06 -0.060
(0.42)
-0.268
(1.19)
-0.105
(0.82)
0.128
(0.54)
Head’s school years -0.014
(2.93)
-0.011
(1.83)
-0.011
(2.73)
-0.011
(1.09)
Age of head -0.003
(1.79)
-0.003
(0.65)
-0.002
(1.65)
-0.002
(0.34)
Food crop income 0.098
(1.77)
0.055
(0.85)
Wage and business -0.054
(1.02)
-0.242
(4.75)
-0.037
(0.71)
-0.035
(0.43)
Constant 5.807
(17.3)
6.407
(10.4)
5.945
(22.4)
5.601
(11.7)
Zero slopes
F
-test
F
(11,56)
=56.1
F
(10,47)
=62.8
F
(10,57)
=184.2
F
(9,48)
=43.4
R
2
.425
.497 .631
.635
Protein
ln PCE 0.643
(19.4)
0.373
(3.03)
0.747
(20.7)
0.629
(10.1)
ln household size -0.179
(4.51)
-0.270
(2.67)
-0.076
(1.92)
-0.276
(2.65)
rf15+ -0.027
(0.21)
-0.158
(0.74)
0.009
(0.08)
-0.224
(1.31)
rf714 0.137
(0.96)
-0.563
(1.89)
0.004
(0.03)
-0.196
(0.81)
rf06 0.424
(2.11)
0.125
(0.32)
0.189
(1.08)
0.649
(1.70)
rm714 0.182
(1.31)
-0.623
(1.76)
0.173
(1.63)
-0.215
(0.61)
rm06 -0.079
(0.53)
-0.326
(0.91)
-0.140
(0.97)
0.475
(1.43)
Head’s school years -0.002
(0.30)
0.002
(0.39)
-0.007
(1.52)
-0.014
(1.89)
Age of head -0.002
(1.16)
-0.005
(0.88)
-0.003
(1.92)
-0.003
(0.59)
Food crop income 0.188
(3.09)
0.037
(0.48)
Wage and business -0.003
(0.05)
-0.248
(2.00)
-0.122
(1.70)
-0.263
(1.77)
Constant -0.109
(0.43)
2.484
(2.72)
-0.237
(1.05)
1.074
(2.17)
Zero slopes
F
-test
F
(11,56)
=68.8
F
(10,47)
=28.8
F
(10,57)
=23.6
F
(9,48)
=4.7
R
2
.533
.508 .689
.695
Note: The sample is N=830 in the rural sector and N
=314 in the urban sector. The reported absolute
t-
values are corrected for the clustered, stratified, and weighted nature of the sample. Variables beginning
with r are demographic ratios, so that e.g., rf714 is the ratio of females aged 7-14 to total house
hold
members. The omitted group is male adults. In the rural regressions there are three economic activity
groups, with households whose head’s main income is from tree crops omitted, while in the urban sector
the omitted group is households whose head’s m
ain income is not from wages or a formal business. The
within cluster regression contains 46 dummy variables for the urban sector and 72 for the rural sector.
9
References
Chen, L., Chowdhury, A., and Huffman, S., 1980. ‘Anthropometric assessment of energy-protein
malnutrition and subsequent risk of mortality among preschool aged children’,
American
Journal of Clinical Nutrition
, 33(12): 1836-1845.
Garrett, J. and Ruel, M. 1999. “Are determinants of rural and urban food security and nutritional
status different? Some insights from Mozambique”,
World Development
27(11):
1955-1975.
Gibson, J. and Rozelle, S. (1998) Results of the household survey component of the 1996 poverty
assessment for Papua New Guinea. Population and Human Resources Division, The
World Bank, Washington DC.
Gibson, J. (1999) Can women’s education aid economic development? The effect on child
stunting in Papua New Guinea”
Pacific Economic Bulletin
14(2): 71-81.
Grantham-McGregor, S., Walker, S., Himes, J, and Powell, C., 1997. ‘Stunting and mental
development in children’,
Nutrition Research
, 16(11): 1821-1828.
Heywood, Peter, Nicola Singleton, and Jay Ross. (1988) “Nutritional status of young children:
the 1982/83 National Nutrition Survey”
Papua New Guinea Medical Journal
31: 91-101.
Shetty, P.S. and W.P.T. James. (1994) “Body Mass Index: A Measure of Chronic Energy
Deficiency in Adults”
FAO Food and Nutrition Paper
56, Food and Agriculture
Organisation, Rome.
Subramanian, Shankar and Angus Deaton, 1996, The demand for food and calories,
Journal of
Political Economy
104(1): 133-162.
World Bank (1999)
Papua New Guinea: Poverty and Access to Public Services
mimeo, The
World Bank, Washington DC.
... Research on the relationship between cash income and food security in PNG indicates a positive correlation between the two (Bourke 2001;Allen and Bourke 2001;Gibson 2001). During the 1997/1998 drought and frost in PNG, villagers' response strategies included the purchase of imported food and the sale of assets to fund food purchases, meaning that the impact was most severe in areas where cash incomes were very low (Bourke 1999). ...
Article
Food security on Malo Island in Vanuatu is examined. All trade between Malo and the outside world passes through a beach on the neighboring island. Data collected there, and on Malo itself, during a long period of fieldwork in 1997—with a short follow-up in 2007—are used to describe the island’s food system qualitatively and in terms of energy availability. The data indicate that 20 % of calories come from food imports, which could be easily substituted with surplus subsistence production in most years. The food system is then analyzed in terms of food security, with consideration given to past and present food systems in the context of economic and climatic variability. The contemporary food system is found to be not only resilient, but far more so than that which existed prior to the commencement of sustained contact with Europeans from around the turn of the twentieth century. While there is some localized pressure on land caused by the dual drivers of population growth and extensive cash cropping, Malo people have been finding innovative solutions that have adapted “traditional” practices and institutions. These findings demonstrate that not all Pacific Islands fit the portrayal of the Pacific as an undifferentiated region characterized by vulnerability and food insecurity. They also demonstrate the importance of social resilience, in this case the adaptive capacity of traditional practices and institutions, to the sustainability of social–ecological systems. The article concludes with reflections on the policy and ethical challenges posed by the “Pacific food insecurity narrative.”
... Where: Y = Mean production of sago (tonnes/person/year) C = Mean energy consumption per year (2665 kcal/person/day × 365 days/year) E = Proportion of energy derived from the staple foods (85%) S = Energy value of sago starch (3570 kcal/kg) The mean energy consumption of 2665 kilocalories per person per day in PNG follows Gibson (2001b), and is similar to the mean value of the recommended daily energy intake of 2600 kcal/day for moderately active 'reference' men and women (WHO 1979:30). The proportion of food energy derived from staple foods (85%) is the mean of ten values in the literature for locations where sweet potato is the dominant staple food (Bourke 1985:92). ...
Book
Full-text available
Estimates of the quantity of food produced in Papua New Guinea (PNG) are important for planning and research. Information contained in the Mapping Agricultural Systems of PNG database of the proportion of land devoted to various food crops allows new estimates to be made. To calculate annual production for each staple crop, estimates of the proportion of garden area devoted to each crop for each agricultural system were combined with census data on the rural population in 2000 (4.3 million rural villagers), mean garden area planted per person per year and mean crop yield for that environment. These data allow estimates to be made for banana, cassava, coconut, potato, Queensland arrowroot, rice, sago, sweet potato, and various taro and yam species. Total annual staple crop production is estimated as 4.5 million tonnes (1050 kg/person/year), with an energy value of 4.3 × 10 12 kilocalories (2770 kcal/person/day). It would require an additional 1.2 million tonnes per year of imported rice to replace the energy value of these staple foods, with a retail value of K2850 million. Sweet potato accounts for almost two-thirds (64%) of production of staple food crops by weight and 63% by food energy. No other staple in PNG contributes more than 10% by weight or food energy. The contribution by weight for the more important foods is: banana (9.7%), cassava (6.0%), yam (6.0%), Chinese taro (Xanthosoma ) (5.0%), Colocasia taro (5.0%), coconut (2.2%) and sago (1.8%). The proportions of food energy produced by these crops are similar to their values for weight. Sago, coconut and rice are exceptions; their proportions of food energy are greater than their contribution by weight. The significance of the various staples varies between provinces in the lowlands, but sweet potato dominates production in the highlands. An independent check of the production figures is made by calculating the energy in the food grown in representative agricultural systems and for all provinces. The energy needs from staple foods of the human population are compared with the available energy. Estimates of production and human needs are very similar for lowland agricultural systems and for provinces where most systems are in the lowlands. For highland provinces, production is 47% greater on average than the estimated human requirements. This is expected and consistent with the large domestic pig populations which are fed significant amounts of sweet potato. These figures suggest that about one third of all sweet potato tubers are fed to pigs in the central highland valleys, or about one quarter of all sweet potato production in PNG. Average sweet potato production per person for all PNG is 670 kg/year, with about 500 kg of this produced for people and the remainder for pig consumption. The figures generated here are compared with estimates made as part of the Survey of Indigenous Agriculture conducted in 1961–1962, with data generated from the 1996 PNG Household Survey, and with figures published annually by the Food and Agriculture Organization of the United Nations. The most important difference between the surveys is that the new estimate for sweet potato production is more than twice that made in the first two surveys. A major change over the past 40 years has been the increased significance of crops of New World origin, that is, cassava, potato, sweet potato and Chinese taro. In contrast, production of staple crops of Asia-Pacific origin (banana, sago, taro and yam) has either decreased or is similar in magnitude to that of 40 years ago.
Article
Full-text available
Papua New Guinea (PNG) is a culturally, environmentally and ethnically diverse country of 7.3 million people experiencing rapid economic development and social change. Such development is typically associated with an increase in non-communicable disease (NCD) risk factors. Aim To establish the prevalence of NCD risk factors in three different regions across PNG in order to guide appropriate prevention and control measures. Methods A cross-sectional survey was undertaken with randomly selected adults (15–65 years), stratified by age and sex recruited from the general population of integrated Health and Demographic Surveillance Sites in West Hiri (periurban), Asaro (rural highland) and Karkar Island (rural island), PNG. A modified WHO STEPS risk factor survey was administered along with anthropometric and biochemical measures on study participants. Results The prevalence of NCD risk factors was markedly different across the three sites. For example, the prevalences of current alcohol consumption at 43% (95% CI 35 to 52), stress at 46% (95% CI 40 to 52), obesity at 22% (95% CI 18 to 28), hypertension at 22% (95% CI 17 to 28), elevated levels of cholesterol at 24% (95% CI 19 to 29) and haemoglobin A1c at 34% (95% CI 29 to 41) were highest in West Hiri relative to the rural areas. However, central obesity at 90% (95% CI 86 to 93) and prehypertension at 55% (95% CI 42 to 62) were most common in Asaro whereas prevalences of smoking, physical inactivity and low high-density lipoprotein-cholesterol levels at 52% (95% CI 45 to 59), 34% (95% CI 26 to 42) and 62% (95% CI 56 to 68), respectively, were highest in Karkar Island. Conclusion Adult residents in the three different communities are at high risk of developing NCDs, especially the West Hiri periurban population. There is an urgent need for appropriate multisectoral preventive interventions and improved health services. Improved monitoring and control of NCD risk factors is also needed in all regions across PNG.
Article
Full-text available
Although recent research revealed an impact of westernization on diversity and composition of the human gut microbiota, the exact consequences on metacommunity characteristics are insufficiently understood, and the underlying ecological mechanisms have not been elucidated. Here, we have compared the fecal microbiota of adults from two non-industrialized regions in Papua New Guinea (PNG) with that of United States (US) residents. Papua New Guineans harbor communities with greater bacterial diversity, lower inter-individual variation, vastly different abundance profiles, and bacterial lineages undetectable in US residents. A quantification of the ecological processes that govern community assembly identified bacterial dispersal as the dominant process that shapes the microbiome in PNG but not in the US. These findings suggest that the microbiome alterations detected in industrialized societies might arise from modern lifestyle factors limiting bacterial dispersal, which has implications for human health and the development of strategies aimed to redress the impact of westernization. Copyright © 2015 The Authors. Published by Elsevier Inc. All rights reserved.
Article
Full-text available
Undernutrition of children 0–60 months old in Mozambique is much higher in rural than in urban areas. Food security is about the same, although substantial regional differences exist. Given these outcomes, we hypothesized that the determinants of food security and nutritional status in rural and urban areas of Mozambique would differ as well. Yet we find that the determinants of food insecurity and malnutrition, and the magnitudes of their effects, are very nearly the same, although some differentiation in determinants of undernutrition does begin to appear among children 24–60 months old. The difference in observed outcomes appears primarily due to differences in the levels of critical determinants rather than in the nature of the determinants themselves.
Article
Stunting is the nutritional indicator most consistently correlated with children's mental development. In Third World countries stunting is usually associated with poor development in young children, and delayed neurosensory integration, low IQ and school achievement in older children. Moreover, stunting in young children predicts poor later development. Generally, when social background is controlled for, the association between stunting and poor development remains. In a recent Jamaican study, nutritional supplementation given to stunted children for 2 years produced an improvement in psychomotor development. Supplementation and linear growth had shared and independent effects on change in development. There was also a significant relationship between change in development and growth over the 2 year period. It is therefore probably that at least part of the poor development found in stunted children is due to poor nutrition.
Article
The authors investigate nutrition and expenditure in rural Maharashtra in India. They estimate that the elasticity of calorie consumption with respect to total expenditure is 0.3-0.5, a range that is in accord with conventional wisdom. The elasticity declines only slowly with levels of living and is far from the value of zero suggested by a recent revisionist literature. In these Indian data, the calories necessary for a day's activity cost less than 5 percent of the daily wage, which makes it implausible that income is constrained by nutrition rather than the other way around. Copyright 1996 by University of Chicago Press.
Article
This paper examines the usefulness of various anthropometric classification systems of nutritional status in prognosticating the subsequent risk of mortality among 2019 children aged 13 to 23 months residing in a rural area of Bangladesh. The indices investigated included: weight-for-age; weight-for-height; height-for-age; arm circumference-for-age; arm circumference-for-height; weight quotient; and height quotient. Cross-sectional anthropometry was conducted during October 1975 to January 1976 and the mortality experience of the study children was followed prospectively over 24 months. Results indicated that severely malnourished children, according to all indices, experienced substantially higher mortality risk. Normal, mild, and moderately malnourished children all experienced the same risk. All indices were found to discriminate mortality risk; weight/age and arm circumference/age were strongest and weight/height weakest. For each index, a threshold level was noted below which mortality risk climbed sharply. The discriminating power of anthropometry was enhanced when maternal weight, maternal height, or housing size were included.
Nutritional status of young children: the 1982/83 National Nutrition Survey” Papua New Guinea Medical JournalBody Mass Index: A Measure of Chronic Energy Deficiency in Adults” FAO Food and Nutrition Paper 56, Food and Agriculture Organisation The demand for food and calories
  • Heywood
  • Nicola Peter
  • Jay Ross Singleton
  • P S Shetty
  • W P T James
Heywood, Peter, Nicola Singleton, and Jay Ross. (1988) “Nutritional status of young children: the 1982/83 National Nutrition Survey” Papua New Guinea Medical Journal 31: 91-101. Shetty, P.S. and W.P.T. James. (1994) “Body Mass Index: A Measure of Chronic Energy Deficiency in Adults” FAO Food and Nutrition Paper 56, Food and Agriculture Organisation, Rome. Subramanian, Shankar and Angus Deaton, 1996, The demand for food and calories, Journal of Political Economy 104(1): 133-162. World Bank (1999) Papua New Guinea: Poverty and Access to Public Services mimeo, The World Bank, Washington DC.
Papua New Guinea: Poverty and Access to Public Services mimeo
  • World Bank
World Bank (1999) Papua New Guinea: Poverty and Access to Public Services mimeo, The World Bank, Washington DC.
Results of the household survey component of the 1996 poverty assessment for Papua New Guinea. Population and Human Resources Division
  • J Gibson
Gibson, J. and Rozelle, S. (1998) Results of the household survey component of the 1996 poverty assessment for Papua New Guinea. Population and Human Resources Division, The World Bank, Washington DC.