Implication of new WHO growth standards on identification of risk factors and estimated prevalence of malnutrition in rural Malawian infants

Background: The World Health Organization (WHO) released new Child Growth Standards in 2006 to replace the current National Center for Health Statistics (NCHS) growth reference. We assessed how switching from the NCHS to the newly released WHO Growth Standards affects the estimated prevalence of wasting, underweight and stunting, and the pattern of risk factors identified. Methodology/Principal Findings: Data were drawn from a village-informant driven Demographic Surveillance System in Northern Malawi. Children (n = 1328) were visited twice at 0-4 months and 11-15 months. Data were collected on the demographic and socio-economic environment of the child, health history, maternal and child anthropometry and child feeding practices. Weight-for-length, weight-for-age and length-for-age were derived in z-scores using the two growth references. In early infancy, prevalence estimates were 2.9, 6.1, and 8.5 fold higher for stunting, underweight, and wasting respectively using the WHO standards compared to NCHS reference (p<0.001 for all). At one year, prevalence estimates for wasting and stunting did not differ significantly according to reference used, but the prevalence of underweight was half that with the NCHS reference (p<0.001). Patterns of risk factors were similar with the two growth references for all outcomes at one year although the strength of association was higher with WHO standards. Conclusions/Significance: Differences in prevalence estimates differed in magnitude but not direction from previous studies. The scale of these differences depends on the population's nutritional status thus it should not be assumed a priori. The increase in estimated prevalence of wasting in early infancy has implications for feeding programs targeting lactating mothers and ante-natal multiple micronutrients supplementation to tackle small birth size. Risk factors identified using WHO standards remain comparable with findings based on the NCHS reference in similar settings. Further research should aim to identify whether the young infants additionally diagnosed as malnourished by this new standard are more appropriate targets for interventions than those identified with the NCHS reference. © 2008 Prost et al. The study was carried out in the southern part of Karonga District, northern Malawi, between August 2002 and October 2004. Out of 1,588 live births recorded in the DSS in the study area, 122 (7.7%) infants either died or left the study area before the follow-up visit, and 50 (3.1%) were excluded for being twins. The analysis of risk factors for malnutrition and prevalence estimates for wasting, underweight and stunting at follow-up included 1328 infants (83.6%) after excluding 88 (5.5%) who had their follow-up visit later than 15 months after birth. The prevalence estimates of underweight and stunting at baseline, when first seen after birth were derived from 1205 infants (75.9%) after excluding 123 (7.7%) infants registered and first assessed more than 120 days after birth. The baseline prevalence estimates of wasting were conducted on 1148 (72.3%) infants after excluding 57 (3.6%) infants with baseline length<49 cm, as the NCHS growth reference is not suitable for calculation of the weight-for-length index for smaller children. Exclusions are summarized in figure 1 . Footnote: †The NCHS reference does not allow for calculation of weight-for-length for children <49 cm Background information on dwelling characteristics, demographic and socio-economic data were drawn from a house-to-house census implemented by trained staff using a standard protocol at the launch of the DSS from August 2002 [13]. Vital events were notified by village informants each responsible for 15–60 households. Notified births were followed by a baseline visit by a project interviewer to formally register the birth and to record the mother's and infant's anthropometric measures and information on feeding practices, health and immunization. A follow-up visit was scheduled 12 months after the birth registration to reassess the child and mother's nutritional status as well as feeding practices, health and immunization. The main caregiver was asked for the approximate age in months when different types of food and beverages were introduced; median age at interview was 12 months (range 11 to 15 months). Throughout the analysis, the term “introduction of water” refers to water and water-based beverages. Complementary food includes breast milk substitutes, cow's milk and maize–based weaning porridges (vernacular: dawale for thin porridge and bara for thick porridge). Family food is defined as all other food items including juices and solid foods. Weight was measured using a spring scale (100 g increments) and length was measured supine using graduated polyurethane plastic mats (0.5 mm increments). Nutritional indices were derived as Z-scores at both time points using the WHO standard and the NCHS reference. Z-scores represent the difference between the height or weight of a child and the median height or weight of the reference population (for the same age and sex) divided by the standard deviation of the reference population. Global wasting, stunting and underweight were defined as weight-for-length, length-for-age and weight-for-age <-2 z-scores respectively. Maternal nutritional status was assessed using the mid-upper-arm-circumference (MUAC), measured using steel tape (1 mm increments). There is no consensus over the use of MUAC for the classification of adult nutritional status. Cut-offs ranging from 18.5 cm [14] to 22 cm [15] have been proposed to define undernutrition. In our analysis, we used a conservative 21 cm cut-off under which the MUAC has been associated with a Body Mass Index<16 kg/m2 in adult women [16], which is widely used by relief agencies for enrolling pregnant and lactating mothers into supplementary feeding programs. Data were double-entered in MS Access 97. The plausibility of measurements was checked electronically at the point of data entry and implausible values were referred back to the field for confirmation [13]. Calculation of nutritional indices with reference to the WHO standards was done in STATA v.9.2 (StataCorp Ltd, Texas, USA) using a macro provided by WHO [17]. Calculation in reference to the NCHS reference was performed in EpiInfo v.6.04d (Center for Disease Control and Prevention, Georgia, USA). The software manufacturers' default settings were applied regarding cut-offs for biologically improbable values ( Table S1 ). Out of range values of z-scores were recoded as missing. The analyses of risk factors for malnutrition were performed in STATA v.9.2 using logistic regression. The three nutritional indices at 11–15 months were the outcomes. Independent variables with even weak evidence of a crude association (p<0.1) with one of the 3 outcomes in the univariate analysis were eligible for inclusion in the multiple logistic regression analysis, as well as variables that have been identified as risk factors for at least one of the outcomes in other local studies [18], [19], [20]. Housing conditions were measured using a dwelling score based on materials used for building the dwelling. The value of household assets was scored to classify the households into four broad categories of “wealth”. Age at the follow-up interview and sex were kept in the model a priori. Further adjustment for health related variables (vaccination status, hospitalization, history of consulting a traditional healer) was made in final models as these variables could be on the causal pathway between socio-economic variables and malnutrition. Finally, adjustment was made for nutritional status at baseline interview (excluding wasting to avoid co-linearity). The strength of the statistical association was assessed using Wald's test and investigation of potential interactions was performed using Likelihood Ratio Test. The statistical significance of the difference between the prevalence estimates of each outcome calculated with both growth references was assessed using McNemar's z-test. Multivariate models were built for each of the 3 outcomes calculated with the WHO standards, using a forward stepwise technique. The selected risk factors were then incorporated in models using the outcomes calculated with the NCHS reference. For each outcome, the direction and strength of the associations were then compared between models. The Karonga DSS called the “Continuous Registration System” was granted ethical approval from the National Health Sciences Research Committee of Malawi and the London School of Hygiene and Tropical Medicine Ethics Committee. Heads of participating households gave verbal consent for being included in the DSS. A further application to use the DSS data for the present study was granted approval from the London School of Hygiene and Tropical Medicine Ethics Committee. The committee accepted the initial verbal consent since it would have been impossible to get a written consent from the guardians of each individual infant included in this analysis due to vital events and population movements.