|
Longitudinal Growth Patterns
of Pakistani Infants in a Clinic Based Growth Promotion Program
|
Shamim A. Qazi, Mushtaq A. Khan, Talat Rizvi,
Zubaida Khatoon, *Karen E. Peterson
From the Children Hospital, Pakistan Institute of
Medical Sciences, G-8/3, Islamabad, Pakistan, and *Departments of
Maternal and Child Health and Nutrition, Harvard School of Public
Health, Boston, Massachusetts 02115, USA.
Correspondence to: Dr. Shamim A. Qazi, Department of
Child and Adolescent Health and Development, World Health
Organization, 20 Avenue Appia, Geneva 27, CH 1211 Switzerland.
E-mail: qazis@who.int
Manuscript received: March 15, 2002, Initial review
completed: April 23, 2002,
Revision accepted: July 30, 2003.
Abstract:
Objective: To describe the operation of
growth monitoring and promotion (GMP) clinic and longitudinal growth
patterns of children. Design: Prospective observation and
intervention. Setting: Outpatient department of a teaching
hospital. Methods: Less than 6 months old infants were
registered at GMP clinic and followed for up to two years of age.
Mothers were provided information, education and counselling about
healthy growth of their infants. The outcome measure was change in
weight after follow-up. Results: We enrolled a cohort of 553
children in the first 6 months of life, of which 318 (57.5%) were
males. Mean follow up period was 15.7 (SD ± 6.4) months. At
enrollment 207 (37.4%) were underweight [weight-for-age Z score (WAZ)
< –2], of which 153 (73.9%) were from low-income families. The
children from poorer families also gained weight regularly, although
it was less than the higher income families. Of 346 infants with
normal weight at registration, 305 (87.6%) maintained their weight
gain by last follow-up visit. Of 207 underweight infants at
enrollment 128 (62.4%) improved their weight gain by the last
follow-up visit. Overall 433(81.0%) infants followed-up at our GMP
clinic either maintained or improved their weight gain.
Conclusions: Children at risk of undernutrition should be
identified at an early age and through effective interaction between
health workers and the family, their growth can be improved.
Children of poor families can also benefit from this activity,
provided a comprehensive approach is made available.
Key words: Growth monitoring, Hospital based,
Infant, Undernutrition.
Child undernutrition is still a major public
health problem in Pakistan in the 1990’s, despite an increase of the
gross national product (GNP) per capita and substantial reduction in
urban and rural poverty in recent years (1). Only a few countries
worldwide have a higher prevalence of low birth weight infants,
estimated at 25% in 1988 (1,2). According to survey finding among
children under five years of age, 40.4% were underweight in 1990-91,
50.0% were stunted and 9.2% showed evidence of wasting(3). Reduction
of moderate malnutrition to 20% has been the goal and focus of several
Government of Pakistan (GOP) nutrition programs and policies,
consistent with the commitments of international agencies to halve the
proportion of child undernutrition worldwide(3-6).
Analyses of risk factors for child undernutrition
in Pakistan suggest that improving food security alone is unlikely to
advance the overall nutritional status of children(7). The nutrition
strategies must be linked to primary health and development programs
that improve breast-feeding and weaning practices, decrease the
incidence of infectious disease, increase levels of maternal
education, and improve utilization of family planning services(7). GOP
nutrition strategies include integration of nutrition related
activities into existing primary health care services(2).
Initiating growth monitoring and promotion (GMP)
activities early in life is essential because prevalence of
malnutrition generally does not increase after 12-18 months of
age(8,9). Studies of Pakistani children have shown that weight gain in
children of lower socioeconomic classes was reduced in the first 12
months of life and length gain was lower than expected from 6 to 18
months of age (10).
Isolated efforts made to develop malnutrition
rehabilitation programs in pediatric facilities in Pakistan(11) in
most cases identified malnourished infants when they came either for
routine "well child care" or "sick child care". Emphasis was placed on
reversing faltering growth by treating illnesses, nutrition education
and occasionally food supplements. Acute treatment was largely
unsuccessful in sustaining the rehabilitated nutritional status and
access to treatment for eligible children was also low. To avoid those
limitations, we adopted an alternative approach of identifying and
targeting children in first six months of life for preventive,
promotive and preemptive approaches to prevent and reduce
mal-nutrition. The objectives of this manuscript are to (i) describe
the operation of a clinic-based growth promotion program in a
Pakistani setting; and (ii) document growth patterns observed over the
first 24 months of life.
Subjects and Methods
Institution
The Children’s Hospital, Pakistan Institute of
Medical Sciences has 200 beds and it treats approximately 400 patients
daily in outpatient and emergency departments. Majority of children
are brought directly to the hospital by parents or family members and
only a small proportion is referred by community general physicians.
The Growth Monitoring and Promotion (GMP) clinic was established in
1988 in the primary health care (PHC) block. The primary health care (PHC)
block also houses clinical service delivery clinics, immunization
clinic, family welfare center (FWC) providing family planning
services, lactation management clinic helping mothers with lactation
problems, oral rehydration therapy (ORT) unit and acute respiratory
infections (ARI) control unit. The GMP clinic is staffed with health
education assistant, a graduate trained in health education,
nutri-tional anthropometry, nutrition education, communication,
standardization of instru-ments and PHC strategies mentioned above.
She was assisted by an experienced aya (a paramedical worker) with
eight grades education, trained in nutritional anthro-pometry,
preparation and demonstration of complementary foods preparation and
communication with the mothers. They were trained and supervised by a
public health physician and a preventive pediatrics physician.
Continuous supervision and regular on-the-job refresher-training were
conducted.
GMP clinic protocol
Children less than six months of age coming to the
OPD for any of the above services were referred to the GMP clinic for
assessment and education (Fig 1). Initial assessment included weight,
nutritional history by the GMP clinic staff and a physical examination
by primary care physician in the OPD. Zero error was corrected before
weighing the child on a tray type scale (Tanita, Japan) with 50-gram
divisions. Accuracy of the scale was checked weekly by known weights.
Mothers were given advice by GMP staff on breast feeding,
individualized specific nutrition issues, immunization and child
spacing and referred for specific services when appropriate. For low
birth weight children mothers were given advice to keep the baby warm,
feed more frequently and to bring the child immediately if the child
fell sick. Mothers were referred for assessment of other risk factors
leading to low birth weight children. Simple complementary foods for
demonstration purposes were cooked by aya for children and fed by the
mothers. The first assessment at registration would take approximately
30 minutes and the subsequent visits would take on an average 15
minutes. Feeding demonstration would take approximately 45 minutes.
Fig 1. Growth Monitoring and Promotion (GMP)
Clinic at the Children's Hospital, Islamabad.
Growth chart
In the nutrition rehabilitation clinic prior to
1988 we used unisex growth charts based on the World Health
Organization (WHO) guidelines that depicted different nutritional
status of less than five years old children by color schemes. Since
growth faltering occurs most frequently in children between 6-18
months of age (12-14), we adapted the under five year old growth
charts for children 0-24 months. Our chart only had two curves, based
on upper two lines of the WHO growth chart(15). For more precision in
tracking growth we used 250 g divisions for weight.
Data collection
Children were registered for growth monitoring and
promotion activities in GMP clinic and an effort was made to follow
all children for two years from the date of registration. Information
about birth weight (reported by mother, low birth weight (LBW) defined
as birth weight less than 2500 g), perceived birth size (small or
normal as perceived and reported by mother), feeding practice (breast,
bottle, mixed, other), reported monthly cash family income (less than
Pak rupees 3000 defined as low income), and number and age of siblings
was recorded for each child by GMP staff during first visit. As birth
weight was not available for all infants, mothers were asked about
their perception of infants’ birth size as a proxy. All children were
scheduled for clinic follow-up visits at one-month intervals up to 24
months of age. No home visits were made because of lack of resources.
At each follow-up visit weight was recorded on the growth card, growth
patterns interpreted by health worker and discussed with the mother.
The GMP staff also recorded information about
illness and other problems during follow-up period for each child,
which is not reported here.
Data analysis
Data were entered and cleaned using EPI INFO(16).
Frequencies, descriptive analyses, relative risk (RR) and 95%
confidence intervals were calculated. Nutrition anthro-pometry module
of EPI INFO was used to calculate weight for age z-score. Underweight
was defined as weight-for-age ‘z’ score (WAZ) less than minus two,
normal as WAZ equal or more than minus two(3). Kappa Statistic was
calculated by SPSS (17).
Results
Five hundred and fifty three children were
registered, of which 334 (60.4%) were registered in the first month of
age and 219 (39.6%) between 2-6 months of age. Three hundred and
eighteen (57.5%) were males. Three hundred and seventy five (67.8%)
infants were born in low-income families (Table I). Birth weight
information was available only in 341 children (61.7%) of whom 53
(15.5%) were LBW. One hundred and twenty-five (22.6%) mothers
perceived their baby to be small at the time of birth. The agreement
was good between mothers’ perception of birth size and birth weight
categories of LBW and normal (Kappa= 0.76). Low-income families were
more likely to have small size babies (RR 1.18, 95% CI 1.05-1.33) P =
0.014). Two hundred and seven (37.4%) infants were underweight at
registration. A higher proportion of infants from low-income families
were under- weight compared to infants in moderate- income families
(OR 1.15, 95% CI 1.03-1.29, P = 0.017). There was no statistically
signi-ficant correlation between nutritional status at the time of
registration and number of siblings. Breastfeeding was more common in
low-income families.
Characteristics of Study Children, Stratified by Monthly Income (N=553).
| |
Income in Pakistani rupees* [n(%)] |
|
|
Factors |
Low income
n=375 |
Moderate income
n=178 |
Relative Risk
(95% CI) |
P-value |
| Perceived birth size |
|
|
|
|
Normal (n=428)
|
279 (65.2)
|
149 (34.8)
|
|
|
Small (n=125)
|
96 (76.8)
|
29 (23.2)
|
1.18 (1.95-1.33)
|
0.014
|
| Birth weight† |
|
|
|
|
Normal (n=288)
|
165 (57.7)
|
123 (42.3)
|
|
|
LBW (n=53)
|
37 (69.8)
|
16 (30.2)
|
1.22 (0.99-1.49)
|
0.088
|
| Number of siblings‡ |
|
|
|
|
None (n=223)
|
143 (64.1)
|
80 (35.9)
|
|
|
1-3 (n=247)
|
167 (67.6)
|
80 (32.4)
|
1.05 (0.93-1.20)
|
0.425
|
4 or more (n=83)
|
65 (78.3)
|
18 (21.7)
|
1.22 (1.05-1.42)
|
0.018
|
| Nutrition status at registration§ |
|
|
|
|
Normal (n=346)
|
222 (64.2)
|
124 (35.8)
|
|
|
Underweight (n=207)
|
153 (73.9)
|
54 (26.1)
|
1.15 (1.03-1.29)
|
0.017
|
| Nutrition status at last follow up§ |
|
|
|
|
Normal (n=433)
|
274 (63.3)
|
159 (36.7)
|
|
|
Underweight (n=120)
|
101 (84.2)
|
19 (15.8)
|
1.33 (1.20-1.48)
|
0.0001
|
| Mode of feeding at registration¶ |
|
|
|
|
Breast feeding (n=308)
|
223 (72.4)
|
85 (27.6)
|
|
|
Bottle feeding (N=72)
|
49 (68.1)
|
23 (31.9)
|
0.94 (0.,79-1.12)
|
0.461
|
Breast and bottle feeding (n=141)
|
81 (57.4)
|
60 (42.6)
|
0.79 (0.68-0.93)
|
0.001
|
Breast and cup and spoon (n=20)
|
14 (70.0)
|
6 (30.0)
|
0.97 (0.72-1.3)
|
0.816
|
Cup and spoon (n=9)
|
8 (88.9)
|
1(11.1)
|
1.23 (0.96-1.56)
|
0.272
|
* 1 US $ = Rs. 60.0, Low income <3000.00 and moderate income ³ 3000.00
† Birth weight information available for 341 children
‡ Odds ratios and p-values are in comparison with no siblings data
§ Normal = ³ –2 WAZ, Underweight = < –2 WAZ
¶ For three children feeding information was not available, Odds ratios and p-values
are in comparison with only breastfeeding data
Change in nutrition status at registration and
after last follow-up by birth weight and perceived birth size is shown
in Table II. As expected infants perceived to be small at birth were
more likely to be underweight at registration (RR 4.54, 95% CI 3.
76-5.49, P <0.0001). After last follow-up, only 120 children (21.7%)
remained underweight. As compared breastfed infants, exclusively
bottle fed infants were more underweight at registration (RR 2.15, 95%
CI 1.51-3.07, P = 0.0002).
TABLE II
Birth Weight and Perceived Birth Size Compared with Nutritional Status at
Enrollment and at last Follow up (n=553)
| |
|
Nutrition status*
[n(%)] |
|
|
Parameter |
At registration |
At last follow up |
| |
Normal |
Underweight |
Normal
|
Underweight |
| Birth weight† |
|
|
|
|
Normal (n = 288)
|
226 (78.5)
|
62 (21.5)
|
257 (89.2)
|
31 (10.8)
|
LBW (n = 53)
|
—
|
53 (100.0)
|
34 (64.2)
|
19 (35.8)
|
Perceived birth size
|
Normal (n=428)
|
339 (79.2)
|
89 (20.8)
|
365 (85.3)
|
63 (14.7)
|
Small (n = 125)
|
7 (5.6)
|
118 (94.4)
|
68 (54.4)
|
57 (45.6)
|
* Normal = ³–2 weight for age z score (WAZ), underweight = <–2 WAZ.
† Birth weight information was available only for 341 children.
Figure 2 gives median weight gain by monthly family
income and Figure 3 reports it by gender. Of 346 normal weight infants
at the time of registration, for 305 (87.6%) WAZ remained ³–2 at last
follow-up visit, whereas 41 (16.5%) became underweight (WAZ <–2). Of
207 underweight infants at the time of registration, for 128 (62.4%)
WAZ became ³–2 at last follow-up visit and 79 (38.5%) remained
underweight.
Fig. 2. •Median
weight for all children ▀ Median weight by family income >=Pak.
Rs. 3000;
▲
Median weight by family income < Pak. Rs. 3000.
Fig. 3. • Median weight of male children; ∆Median
weight of female children.
Mean length of follow-up was 15.7 ± 6.4 months (95%
CI 15.1-16.2). Only 227 (41.0) children came for follow-up beyond 18
months of age, whereas 26 (4.7%) children, stopped coming for
follow-up before six months of age. Figure 4 provides graphic
representation of loss to follow-up. A significantly higher proportion
of small birth size infants (10/125) stopped coming for follow-up
between 0-5 months as compared to normal size children (14/428) (RR
1.92,95% CI 1.16-3.16 P = 0.02). After first five months of follow-up
there was no significant difference in those who were followed for a
longer time by birth size, birth weight, family size or family income.
Fig. 4.
Follow-up of children at the GMP clinic
Discussion
Although 37% of our study children were underweight
at the time of registration, most of them showed catch-up growth; and
average weights of small birth size infants improved by four months of
age. Weight gain was maximal in the first few months of life. At the
end of follow-up only 21.7% children were still underweight compared
with the national reported estimate of 35.8%(3).
Our data shows that children from poorer families
had lower median weights than the children from better income families
and females children had lower weights than males, although the growth
patterns were parallel in both groups. Thus, nutrition education and
advice alone was not enough and female children from very poor did
worse as compared to other children. Thus there is a need to reach the
most vulnerable groups through interventions, that may not work best
if targeted at secondary and tertiary care institutions.
Improvement in nutrition status of LBW infants and
sustained nutritional status in normal children may have occurred for
two reasons. First, enrollment for prospective monitoring as soon as
possible after birth was critical, because the age at the peak of
growth faltering is too late and not an optimal time to intervene with
GMP. Interventions must be directed at or before the age when the risk
behaviors/outcomes first begin to act on the child. Catchup growth in
the LBW infants is most pronounced in the first 6-8 months of life, if
nutritional supplies are adequates(18), so they need to be monitored
from birth. The recommended timing and evaluation of growth monitoring
(8,9) often fails to take into account the benefits of its integration
with PHC interventions. Other growth promotive messages (e.g.,
exclusive breast feeding, appropriate use of complementary foods and
feeding during illness), must also be part of the health education.
Moreover, consistent with the national policy of
growth promotion(11), we tried maternal behavioral change through
information, education and communication to achieve improved nutrition
at home and thus improve growth of their children.
Several potential limitations of the data must be
considered in interpreting the results, First, a number of children
were not brought regularly for follow-up, Regular visits place a large
burden particularly on low-income families in terms of travel costs
and time. Furthermore, mothers, especially very poor were less
inclined to bring older children for weighing, as they did not value
the time investment for the visit. In the first five months the
disproportionate loss to follow-up of small birth size infants could
be due to early infant mortality, resulting in a survival bias. We
were unable to evaluate this possibility because of limited resources.
Irregular follow-up, time and other cost limitations have been
reported in other locations using growth-monitoring program(19).
Second, birth weight data was not available for all children. We used
the perceived birth size and found it to be a reasonable proxy for LBW
babies, Third, this was a hospital-based convenience sample, not
necessarily representative of the community, therefore these results
may not be generaliza-ble. Fourth, infections are common in Pakistan
and a drop in growth velocity may occur after illness, resulting in
some normal weight children at registration who became underweight and
some underweight children remained underweight, despite advice to
provide extra nutrition during and after illness to reverse the drop.
Fifth, fewer female children, were registered, consistent with health
facility utilization for females in Pakistan(20). Finally, the study
design did not permit assessment of impact or adequacy of the model,
because there was no control group who was not exposed to our
intervention. Ethically it was inappropriate not to provide
nutritional advice to a control group.
GMP is less likely to succeed if health & care
providers are not trained to assess, interpret and act on the
information provided by weighing(8,21). Our approach was a combined
screening, education and communication strategy where workers were
able to assess, interpret the results and act on it. Such an approach
should be "family centred", by involving parents in decision making
and addressing health and social needs of the family.
Growth monitoring and prospective interventions
require interaction between health workers and the family, a health
approach that is not currently in vogue in Pakistan. Our study
demonstrates that this approach could be an important part of the
delivery of primary health care.
Acknowledgement
The authors are indebted to the following people
who reviewed the manuscript and gave their suggestions: Dr. Patricia
Hibberd, Clini-cal Research Institute, New England Medical Center,
Boston and Dr. Jonathan Simon, Director Center for International
Health, Boston University, Boston, United States.
Contributors: SAQ was responsible for
initialization, institutionalization and overall supervision of the
GMP clinic, conceptualization of the manuscript, data management,
analysis and manuscript writing. MAK helped in initialization and
institutionalization of the GMP clinic, conceptualization of the
manuscript and contribution in manuscript writing. TR was involved in
supervision and management of the GMP clinic on day-to-day basis,
conceptualization of the manuscript and contributed to data analysis
and manuscript writing. ZK helped in managing the GMP clinic on
day-to-day basis and data collection. KEP was involved with
conceptualization and contribution in manuscript writing.
Funding: None.
Competing interests: None stated.
|
Key Messages |
• Children at risk of undernutrition should be identified at
an early age.
• Growth of infants can be improved by an effective
interaction between health workers and the family.
|
|
1. Garcia M, Mason J. Second Report on the
World Nutrition Situation, Volume II: Country Trends and
Statistics. Geneva: United Nations Administrative Committee on
Coordination-Subcommittee on Nutrition (ACC/SCN), 1993.
2. Nutrition Section, Planning and
Development Division, Government of Pakistan. Pakistan Country
Paper, prepared for Joint FAO/WHO International Conference on
Nutrition, Rome, December 1992. Government of Pakistan,
Islamabad 1991.
3. de Onis M, Monteiro C, Akre J, Clugston G.
The worldwide magnitude of protein-energy malnutrition: An
overview from the WHO global database on child growth. Bull WHO
1993; 71: 703-712.
4. Ramalingaswami V. New global perspectives
on overcoming malnutrition. Am J Clin Nutr 1995; 61: 259-263.
5. UN. World Declaration and Plan of Action.
World Summit for Children. New York: United Nations, 1990.
6. Garcia M, Mason J. Second Report on the
World Nutrition Situation, Volume I: Global and Regional
Results. Geneva: United Nations Administrative Committee on
Coordination-Subcommittee on Nutr (ACC/SCN), 1992.
7. Alderman H, Garcia M. Food security and
health security: Explaining the levels of nutritional status in
Pakistan. Econ Dev Cult Change 1994; 42: 485-507.
8. Pearson R. Thematic evaluation of UNICEF
support to growth monitoring. New York: UNICEF, 1993.
9. Huttly S, Victora CG, Bartros FC, Teixeira
A, Vaughan JP. The timing of nutritional status determination:
Implications for interventions and growth monitoring. Eur J Clin
Nutr 1991; 45: 85-95.
10. Karlberg J, Ashraf RN, Saleemi M, Yaqoob
M, Jalil F. Early child health in Lahore, Pakistan: XI. Growth.
Acta Paediatr (suppl) 1993; 390: 8119-8149.
11. Committee on Growth Promotion, Nutrition
Section, Planning and Development Division. Report of the
Committee on Growth Promotion. Islamabad: Government of
Pakistan, 1992.
12. Government of Pakistan. Diarrheal
disorders and feeding practices in Pakistan. Planning and
Development Division. Government of Pakistan, Islamabad. 1984 .
13. Rohde JE. Editorial. Indian J Pediatr (suppl)
1988; 55: 81-82.
14. Rohde JE. Beyond survival: Promoting
healthy growth. Indian J Pediatr (suppl) 1988; 55: 83-88.
15. World Health Organization. The growth
chart: a tool for use in infants and childcare. Geneva, WHO
1986.
16. EPI INFO 6. World Health Organization,
Geneva and Centers for Disease Control, Atlanta. 1995.
17. SPSS 10.0 for Windows. SPSS Inc. Chicago
1999.
18. Peterson KE, Frank DA. Feeding and growth
of premature small-for-gestational-age infants. In:
Taeusch W, Yogman M (editors). Follow-Up management of the high
risk infants. Boston: Little Brown and Co; 1987, p 87-204.
19. Yee V, Zerfas A. Issues in growth
monitoring and promotion: summary. Chevy Chase MD: LTS
manuscript, 1987.
20. Ahmad A, Akhtar A, Alizai S and Zia A.
Gender differential in access to health care for Pakistani
children. UNICEF, Islamabad 1990.
21. Dixon RA. Cost-effectiveness of growth monitoring and
promotion. Lancet 1993; 342-317.
|
