Reliability and Validity of an Adapted and Translated Version of the Mullen Scales of Early Learning (AT-MSEL) in Rural Guatemala.

BACKGROUND: A growing literature base supports the use of tests developed in high-income countries to assess children in low resource settings when carefully translated, adapted, and applied. Evaluation of psychometric properties of adapted and translated measures within populations is necessary. The current project sought to evaluate the reliability and validity of an adapted and translated version of the Mullen Scales of Early Learning (AT-MSEL) in rural Guatelama. METHODS: The reliability and validity of the AT-MSEL in rural Guatemala were analyzed for children ages 0-5 years. RESULTS: Interrater reliability coefficients (ICC = 0.99-1.0) and internal consistency (Cronbach's alpha = 0.91-0.93) were excellent for all subscales. General linear models utilizing paired data showed consistency between standard scores (p < 0.0001). Mean raw scores increased with chronological age, as expected. Across age groups, subscales were significantly, positively correlated with one another (p < 0.05 - < 0.001) with one exception, visual reception and expressive language at the 0-10 month age range (p = 0.43). CONCLUSIONS: The AT- MSEL showed strong psychometric properties in a sample of young children in rural Guatemala. Findings demonstrate that the AT-MSEL can be used validly and reliably within this specific population of children. This work supports the concept that tests developed in high-income countries can be used to assess children in low resource settings when carefully translated, adapted and applied.

Receptive language skills among young children in rural Guatemala: The relationship between the Test de Vocabulario en Imagenes Peabody and a translated and adapted version of the Mullen Scales of Early Learning.

BACKGROUND: Children in low- and lower middle income countries (LMICs) often have poorer language skills compared with children from high-income countries. Limited availability of culturally and linguistically appropriate assessment measures in LMICs, especially for young children, can hinder early identification and prevention efforts. Here, we describe receptive language (RL) skills among young children in rural Guatemala and report on the validity of a translated and culturally adapted developmental measure of RL. METHODS: Children (n = 157; m = 53.6 months, range = 42-68 months) enrolled in a prospective cohort study of postnatally acquired Zika virus infection were administered the Test de Vocabulario en Imagenes Peabody (TVIP) and the RL scale from a translated and adapted version of the Mullen Scales of Early Learning (MSEL). Performance on the TVIP was compared with the Latin American normative sample. Correlational analysis examined the relationship between performance on the TVIP and the MSEL-RL. RESULTS: Mean scores were significantly below the normative sample mean on the TVIP, t(126) = -11.04, p < .001; d = 1.00. Performance on the TVIP among children who passed the practice items (n = 127) was significantly positively associated with performance on the MSEL-RL (r = .50, p < .001), but not significantly associated with age or gender. Older age (p < .0001) and female gender (p = .018) were associated with passing the TVIP practice items. CONCLUSIONS: Delays in RL vocabulary were identified among young children in rural Guatemala on the TVIP. The association between scores on the TVIP and the RL scale of the MSEL provides preliminary support for the construct validity of this translated and adapted version of the MSEL.

Anthropometric proxies for child neurodevelopment in low-resource settings: length- or height-for-age, head circumference or both?

Stunting (<-2 SD of length- or height-for-age on WHO growth curves) is the most used predictor of child neurodevelopmental (ND) risk. Occipitofrontal head circumference (OFC) may be an equally feasible, but more direct and robust predictor. We explored association of the two measurements with ND outcome, separately and combined, and examined if cutoffs are more efficacious than continuous measures in predicting ND risk. Infants and young children in rural Guatemala (n = 642; age range = 0.1-35.9 months) were enrolled in a prospective natural history study, and their ND was tested using the Mullen Scales of Early Learning (MSEL) longitudinally. Length- or height-for-age and OFC-for-age were calculated. We performed age-adjusted multivariable regression analyses to explore the association between 1) length or height and ND, 2) OFC and ND, and 3) both length or height and OFC combined, with ND; concurrently, predictively, and longitudinally, as continuous variables and using WHO z-score cutoffs. Continuous length- or height-for-age and OFC z-scores were more strongly associated with MSEL than the traditional -2 SD WHO cutoff. The combination of height-for-age z-score and OFC z-score was consistently, strongly associated with the MSEL Early Learning Composite concurrently (p-values 0.0004-0.11), predictively (p-value 0.001-0.07), with the exception of the 18-24 months age group which had very few records, and in the longitudinal model (p-value <0.0001-0.004). The combination of continuous length- or height-for-age and OFC shows additional utility in estimating ND risk in infants and young children. Measurement of OFC may improve precision of prediction of ND risk in infants and young children.

Cumulative Febrile, Respiratory, and Gastrointestinal Illness Among Infants in Rural Guatemala and Association With Neurodevelopmental and Growth Outcomes.

BACKGROUND: Infectious disease exposures in early life are increasingly recognized as a risk factor for poor subsequent growth and neurodevelopment. We aimed to evaluate the association between cumulative illness with neurodevelopment and growth outcomes in a birth cohort of Guatemalan infants. METHODS: From June 2017 to July 2018, infants 0-3 months of age living in a resource-limited region of rural southwest Guatemala were enrolled and underwent weekly at-home surveillance for caregiver-reported cough, fever, and vomiting/diarrhea. They also underwent anthropometric assessments and neurodevelopmental testing with the Mullen Scales of Early Learning (MSEL) at enrollment, 6 months, and 1 year. RESULTS: Of 499 enrolled infants, 430 (86.2%) completed all study procedures and were included in the analysis. At 12-15 months of age, 140 (32.6%) infants had stunting (length-for-age Z [LAZ] score < -2 SD) and 72 (16.7%) had microcephaly (occipital-frontal circumference [OFC] < -2 SD). In multivariable analysis, greater cumulative instances of reported cough illness (beta = -0.08/illness-week, P = 0.06) and febrile illness (beta = -0.36/illness-week, P < 0.001) were marginally or significantly associated with lower MSEL Early Learning Composite (ELC) Score at 12-15 months, respectively; there was no association with any illness (cough, fever, and/or vomiting/diarrhea; P = 0.27) or with cumulative instances of diarrheal/vomiting illness alone ( P = 0.66). No association was shown between cumulative instances of illness and stunting or microcephaly at 12-15 months. CONCLUSIONS: These findings highlight the negative cumulative consequences of frequent febrile and respiratory illness on neurodevelopment during infancy. Future studies should explore pathogen-specific illnesses, host response associated with these syndromic illnesses, and their association with neurodevelopment.

A prospective cohort study of head circumference and its association with neurodevelopmental outcomes in infants and young children in rural Guatemala.

Microcephaly, an anthropometric marker of reduced brain volume and predictor of developmental disability, is rare in high-income countries. Recent reports show the prevalence of microcephaly to be much higher in lower resource settings. We calculated the prevalence of microcephaly in infants and young children (n = 642; age range = 0.1-35.9 months), examined trends in occipitofrontal circumference (OFC) growth in the year after birth and evaluated the relationship between OFC and performance on the Mullen Scales of Early Learning (MSEL) in rural Guatemala. Multivariable regression analyses adjusted for age were performed: (1) a model comparing concurrent MSEL performance and OFC at all visits per child, (2) concurrent OFC and MSEL performance by age group, and (3) OFC at enrollment and MSEL at final visit by age group. Prevalence of microcephaly ranged from 10.1% to 25.0%. OFC z-score decreased for most infants throughout the first year after birth. A significant positive association between continuous OFC measurement and MSEL score suggested that children with smaller OFC may do worse on ND tests conducted both concurrently and ∼1 year later. Results were variable when analyzed by OFC cutoff scores and stratified by 6-month age groups. OFC should be considered for inclusion in developmental screening assessments at the individual and population level, especially when performance-based testing is not feasible.

Challenges and lessons learned from the rapid operationalization of a prospective cohort to study the natural history and neurodevelopmental outcomes of postnatal Zika virus infection among infants and children in rural Guatemala.

During the course of the 2015-2017 outbreak of Zika virus (ZIKV) in the Americas, the emerging virus was recognized as a congenital infection that could damage the developing brain. As the Latin American ZIKV outbreak advanced, the scientific and public health community questioned if this newly recognized neurotropic flavivirus could affect the developing brain of infants and young children infected after birth. We report here the study design, methods and the challenges and lessons learned from the rapid operationalization of a prospective natural history cohort study aimed at evaluating the potential neurological and neurodevelopmental effects of postnatal ZIKV infection in infants and young children, which had become epidemic in Central America. This study enrolled a cohort of 500 mothers and their infants, along with nearly 400 children 1.5-3.5 years of age who were born during the initial phase of the ZIKV epidemic in a rural area of Guatemala. Our solutions and lessons learned while tackling real-life challenges may serve as a guide to other researchers carrying out studies of emerging infectious diseases of public health priority in resource-constrained settings.

Sensitive and Prolonged Detection of Dengue Virus RNA in Whole Blood.

Molecular detection of dengue virus (DENV) RNA from serum or plasma provides an accurate acute-phase diagnostic (< 7 days after symptom onset). Detection may be prolonged in whole blood, although data are limited. We tested for DENV by real-time reverse transcription-PCR in 345 paired acute-phase plasma and whole blood samples from individuals with a Flavivirus-like illness in southwestern Guatemala. In 18/18 cases with detectable DENV RNA in plasma, whole blood samples were positive and yielded similar cycle threshold values. In seven individuals with convalescent samples obtained 2-3 weeks later, DENV RNA remained detectable in whole blood but not plasma. In three additional cases, DENV RNA was only detectable in whole blood at the acute visit. In two cases, whole blood detection was linked to a virologically confirmed DENV infection 6-11 weeks earlier. Whole blood DENV RNA detection is sensitive for acute dengue infection and may remain positive for weeks to months.

Parent Report of Health Related Quality of Life in Young Children in Rural Guatemala: Implementation, Reliability, and Validity of the PedsQL in Stunting and Wasting.

In this study, we review the implementation, reliability, and validity of the Pediatric Quality of Life Inventory (PedsQL), a measure of health-related quality of life, in young children in rural Guatemala. Mothers of 842 children (age range = 1-60 months) completed the PedsQL Generic Core Scales 4.0 serially for 1 year. Low (Pearson's r = 0.28, P < .0001) to moderate (Pearson's r = 0.65, P < .0001) consistency in responding over time was shown. The PedsQL did not discriminate reliably between healthy children and those with stunting or wasting. PedsQL scores were not lower during the time of an acute illness. While we found low to moderate evidence for the reliability of the PedsQL in healthy children, it did not discriminate between healthy children and those with stunting, wasting or other acute illness.

Caregiver rating of early childhood development: Reliability and validity of the ASQ-3 in rural Guatemala.

BACKGROUND: Although performance-based assessment of early childhood development is preferred, there are a number of limitations to this methodology in low resource settings (LRSs). Hence, clinicians and researchers often rely on caregiver report screening tools. The Ages and Stages Questionnaire 3 (ASQ) is one of the most widely used caregiver report measures globally. Adequate psychometric properties have been demonstrated in high income settings, especially when used in older children, high- risk children, or those with severe neurodevelopmental delays. However, its utility is more variable within very young children and for use in LRSs. METHODS: The reliability and validity of the ASQ was determined for children ages 0-5 years living in rural Guatemala. Internal consistency and test-retest reliability were assessed, as well as concurrent and predictive validity. Sensitivity, specificity, positive and negative predictive values related to performance-based developmental assessment (Mullen Scales of Early Learning; MSEL) and growth status (i.e. stunting) were also calculated. RESULTS: Internal consistency reliability for the ASQ was adequate, except when results were limited by small sample size. Test-retest reliability ranged from low to moderate (r = 0.08-0.43; p < 0.05-0.01). However, there was significant variability in mean scores over time across ASQ domain scores. In terms of validity, the ASQ did not discriminate adequately between children who performed within or below age-expectations on performance-based developmental testing or those with and without stunting. CONCLUSIONS: The ASQ did not demonstrate adequate psychometric properties in rural Guatemala, consistent with concerns documented in other LRSs. These results indicate that existing caregiver report screening measures of early childhood development should be utilized with caution in LRSs, and alternative methods for assessment or in the development and utilization of caregiver report measures should be considered.

Responding to the Zika Epidemic: Preparation of a Neurodevelopmental Testing Protocol to Evaluate Young Children in Rural Guatemala.

The ongoing Zika virus (ZIKV) epidemic in Latin America presented a unique opportunity to develop a neurodevelopmental assessment protocol for children in a lower middle-income country. Although studies of neurodevelopment in young children have taken place in many diverse global settings, we are not aware of any study that has provided a high level of detail about how a measure was selected and then specifically translated and adapted in a low-resource setting. Here, we describe considerations in measurement selection and then the process of translation and adaptation to assess neurodevelopmental outcomes of infants and young children with postnatal exposure to ZIKV in rural Guatemala. We provide a framework to other research teams seeking to develop similar assessment models across the globe.