Reflexões sobre os parâmetros psicométricos do Inventário Home versão Infant Toddler / Reflections infant Toddler Home Inventory version psychometrics parameters

O ambiente domiciliar e os comportamentos dos cuidadores da criança são considerados fatores relevantes para o desenvolvimento infantil. O Inventário HOME versão Infant Toddler propõe-se a avaliar os aspectos físicos e sociais do contexto familiar e os comportamentos dos cuidadores primários da criança. Este artigo visa descrever o Inventário HOME versão Infant Toddler, considerando as suas características psicométricas, e refletir sobre a sua utilização em pesquisas científicas e seu uso no Brasil. Foi realizada uma descrição da medida considerando as suas características psicométricas, tais como validade, critérios para a construção dos itens e definição do fenômeno avaliado. Com base no processo de reflexão observaram-se lacunas na construção de itens, a falta de padronização do instrumento no Brasil, especialmente na falta de uniformização na tradução do inventário para a língua portuguesa. Apesar disso, considera-se que o instrumento possui vantagens como a fácil aplicação e mostra-se pertinente para o screening do ambiente familiar, auxiliando na criação de programas preventivos adequados para a realidade das famílias estudadas.

The home environment and child carer behaviour are considered relevant factors for childrens' development. Infant Toddler HOME Inventory Version proposes to evaluate physical and social aspects of the family context as well as the behaviour of the primary child carer. This article aims to describe the Infant Toddler HOME Inventory Version, considering its psicometric features and also to think about its use in scientific research in Brazil. To that end a description of the measure was made, taking into account its psychometric features such as validity, criteria for the construction of the items, definition of the evaluated phenomena. Based on these considerations it was observed that there were gaps in the item building, a lack of standardization of the tool in Brazil, particularly in the lack of uniformity of the translation of the inventory questionnaire into the Portuguese language. In spite of this, it was observed that the instrument has advantages such as easy application and that it is pertinent for home environment screening, helping in the creation of preventive programs suited to the reality of the families studied.

Experiencia de la aplicación y criterios para la interpretación de dos versiones del Inventario HOME para infantes de 0 a 3 años de vida / Application experience and criteria for the interpretation of two versions of Infant-Toddler HOME

The Home Observation for the Measurement of the Environment (HOME) Inventory is the most widely used tool in evaluating the environmental stimulation and family potential to provide necessary cognitive, social, and emotional support available to the child development. The HOME uses an observation/interview format to measure specific interactions between the caregiver and the child, various issues and events that might stimulate the child, and a variety of experiences provided at home or by family members that offer opportunity for growth and development. There are several versions to use from birth to age 3 (Infant-Toddler HOME), the most famous data come the late 60's (1968), it is composed by 45 items (HOME-45) clustered into six subscales. It has a back up of 40 years of publications, over 650 papers, associating with the children's language development, intellectual performance, and academic achievement or psychosocial and socioeconomic characteristics of the family. In Mexico, there are only two published version reports of the use of HOME-45. Most reports refer to the version of Cravioto from the late 60's, with 7 subscales and 62 items (HOME-62) from which there's poor information in the literature about the criteria for to interpreter the outcome results. The aim is to show the results of the implementation of both versions of HOME inventory, and provide criteria for interpretation in a sector of the Mexican population of low socioeconomic status. Material and methods We studied 1031 children and their homes from three communities of low socioeconomic status of central Mexico. Explaining the procedures and obtained informed consent, they were visited at home to implement the HOME Inventory and the socio-economic survey, with a 95% interobserver reliability. The procedures performed were approved by the Ethics and Research Committees from National Pediatric Institute. We applied the HOME-45 and HOME-62 in a integrated format to obtain the total score and each subscale from each version. It was also felt cut-off formed three groups with low stimulation (lowest fourth), middle stimulation (middle half) and high stimulation (upper fourth). The HOME-45 applies all the items to children 0 to 3 years; in the HOME-62 before 3 months old are evaluated only 50 items and after 3; 9; 12; 15 and 18 months old apply 54; 55; 57; 59 and 62 items respectively. Depending on the age, 26 to 36 items are common, representing from 58 to 80% of HOME-45 and 52 to 58% of HOME-62. There is no correspondence between the subscales of both versions, they have a different name and composition of items; so, we consider both versions as different. Social survey A modified version of the socio-economic interview of the National Pediatric Institute was applied, therefore three levels were established: medium low, low and very low social-economic status. Statistical analysis Four groups were formed by age range of children, to analyze the version HOME-45 and 6 groups for HOME-62 version, depending on the number of applicable items indicated in the format. The analysis related to HOME-62 considered only those items that the format indicates to apply for specific ages. Analyze was performed using JMP statistical software (version 5.01; SAS). We used ANOVA, Student's t test or Tukey Kramer's test to examine differences between means and chi square test to assess differences between proportions; p< 0.05 was considered statistical significance. Results 52.7% of the children were female, aged 1 to 35 months ( = 16.2 ± 9.3 months) mother was the primary caregiver of the child at 87.8%. The mean mother's age was 26.5 ± 5.7 years with a range of 14 to 43 years, the father's age was 29.8 ± 7.34 years. Mother's education mean was 9 years, 24% elementary school or less. HOME-45 Inventory achieved mean score in the population of 31.02 ± 5.68 points, equivalent to a hits proportion of 0.69 ± 0.13, with limits between 0.51 (variety) to 0.81 (responsibity). We found statistical significance differences in the total score of HOME-45 Inventory according to socioeconomic and marital status, birth order of children, education and age of parents. There were also found differences between all subscales of HOME-45 with the socioeconomic status and parent's education. With scores of HOME-45 Total difference was found between children younger than 6 months respect to older. There were also differences in the same sense in the subscales except for Acceptance and Organization subscales. It was considered the cutoff point for low, medium and high stimulation in the total population (0-35 months), to total scale and each subscale, the most important differences observed occurred in children under 6 months of age. HOME-62 Inventory (Cravioto) The total score and the subscales of HOME-62 also showed associations with sociodemographic variables such as education, age and marital status from parents, socioeconomic family status and birth order of children. Scores reached at the HOME-62 were equivalent to the hits proportion of 0.74 ± 0.1 , all subscales showed mean higher than 0.5 and three above 0.8. There were major differences in the total score of HOME-62 considering the age group proposed, similar differences were observed in most of the subscales. The cut-off points for types of stimulation of total HOME-62 tend to increase with the age of the child. The same happened with the subscales except Frequency and Stability of adult contact and Emotional climate, were descended. It was compared the relationship between the two versions of HOME once the criteria proposed was applied: 719 family environments were similarly characterized by both versions (69.7%) and 235 environments (22.8%) were defined by the HOME-45 with great stimulation while 80 environments (7.8%) were defined by the HOME-62 with great stimulation too. Discussion There are no benchmarks in the Mexican population for any of the two versions of HOME used. And the international benchmarks HOME-45 were obtained from American families for over 30 years, being important to have results related to its application and interpretation of HOME-45 in a sector of the Mexican population, allowing contrasting findings regarding those obtained in other latitudes. HOME-62 version is the most widely used in Mexico, but little has been reported about the achieved scores, because there are no criteria for interpretation. The values proposed in this paper can be used in future comparisons with the results of other studies. Differences in scores of each version of HOME according to the age of the children, make it necessary to show the results adjusted by age ranges, common procedure in several instruments. The results of HOME-45 in the population were similar to those obtained in other data, there were no statistical differences with the Little Rock values in the total score, and it existed with the subscales of Acceptance and Organization. Like in that population, the tendency to increase the total score of HOME-45 with age was due mainly to Learning materials and Variety subscales. The results of HOME-62 provide benchmarks for its interpretation in 6 age group and all subscales. If they are applied to the results of the Cravioto's investigation, they are able to differentiate families with a history of malnourished children in the early years of life. Both versions showed ability to discriminate socio-economic characteristics of family. It is necessary to report results of the implementation of HOME in different contexts in qualitative and quantitative terms for a proper comparison and interpretation of the differences between groups as well as analyzing the relationship between the specific results in the subscales of both versions with the development, behavior and welfare of children.

El instrumento más utilizado en la evaluación del potencial del ambiente familiar para brindar estimulación y apoyo necesario al desarrollo del niño, es el Inventario HOME (Home Observation for the Measurement of the Environment). La versión más conocida internacionalmente, para niños de cero a tres años, contiene 45 reactivos (HOME-45). En México la mayoría de reportes se refiere a una versión modificada por Cravioto con 62 reactivos (HOME-62), con poca información sobre los criterios para su interpretación. No existe correspondencia en el nombre de las subescalas de ambas versiones ni en los reactivos que las conforman, por lo que se consideran ambas versiones como distintas. No existen valores de referencia para la población mexicana de ninguna de las dos versiones utilizadas del HOME. Objetivo Mostrar los resultados de la aplicación de ambas versiones del inventario HOME y ofrecer criterios para su interpretación en un sector de la población mexicana de condición socioeconómica baja. Material y métodos Se estudiaron 1031 niños y sus hogares mediante visitas domiciliarias en tres comunidades del estrato socioeconómico bajo del centro de México. Los procedimientos tuvieron una confiabilidad interobservador del 95% y fueron aprobados por los Comités de Investigación y Ética del Instituto Nacional de Pediatría. Se obtuvo la puntuación para cada una de las versiones y se estimaron puntos de corte formándose tres grupos de estimulación, según la distribución cuartilar. Resultados El cincuenta y dos punto siete por ciento de los niños fue del sexo femenino, o = 16.2 ± 9.3 meses de edad, el cuidador principal del niño fue la madre en el 87.8%. La puntuación alcanzada por la población en el Inventario HOME-45 fue de 31.02 ± 5.68 puntos. Se encontraron diferencias significativas en la puntuación del HOME-45 y las subescalas según el orden de nacimiento del niño, la condición socioeconómica, el estado conyugal, la edad y escolaridad de los padres. Se estimó el punto de corte para la baja, media y alta estimulación observándose las diferencias más importantes en los menores de seis meses de edad. Inventario HOME-62 La puntuación total y la de las subescalas del HOME-62 también mostraron asociación con las variables sociodemográficas; la puntuación total del HOME-62 fue equivalente a una proporción de aciertos de 0.74±0.1. Se observaron diferencias en la puntuación total y en la mayoría de las subescalas del HOME-62 según los intervalos de edad propuestos. Los puntos de corte para los tipos de estimulación del HOME-62 tienden a incrementarse con la edad del niño e igual sucede con la mayoría de las subescalas. La relación entre ambas versiones del HOME, una vez aplicados los criterios propuesto, mostró 69.7% de ambientes familiares tipificados de igual manera por ambas versiones, 22.8% tipificados con mayor estimulación por el HOME-45 y 7.8% de ambientes tipificados con mayor estimulación por el HOME-62. Conclusiones Es importante contar con referentes para la interpretación del HOME-45 en habitantes de México, que permita contrastar los hallazgos respecto a los obtenidos en otras latitudes. El HOME-62 es la versión más utilizada en México, pero se han reportado poco las puntuaciones alcanzadas, en parte porque no hay criterios para su interpretación. Se presentaron los resultados ajustados a intervalos de edad, procedimiento común en diversos instrumentos. Los valores propuestos en el presente trabajo pueden usarse en futuras comparaciones con los resultados de otros estudios. Los resultados del HOME-62 aportan referentes para su interpretación en seis intervalos de edad en cada subescala. Ambas versiones muestran capacidad para discriminar características socioeconómicas de las familias. Es necesario reportar resultados de la aplicación del HOME en diversos contextos en términos cualitativos y cuantitativos para una adecuada comparación e interpretación de las diferencias entre grupos, así como analizar las relaciones entre los resultados específicos en las subescalas de ambas versiones con el desarrollo, comportamiento y bienestar del niño.

Consistencia interna a lo largo de un año del Inventario HOME-infantes en un grupo de niños de la Ciudad de México y zona metropolitana

resumen está disponible en el texto completo

Summary As the environment is a crucial source in an infant's development, it is important to assess the proximal environment where a variety of social relationships take place. Experiences derived from the home environment allow the specific activities that a child builds actively. These opportunities have an outstanding impact on an infant's development. The home concept and its influence on development led to the discovery and interrelations of several dimensions and today's emphasis relies on identifying the mechanisms of the variability of environment that exert an influence on the variability of development. Under this perspective, at the end of the 1960's the Infant/Toddler HOME Inventory was developed by Caldwell and Bradley. Very soon, Cravioto adapted it for its use in a Mexican population according to the characteristics of typical families and cultural aspects. The purpose of this investigation was to establish the internal consistency of the Infant/Toddler HOME Inventory in the version adapted by Cravioto through the assessment of the precision of its scoring in terms of internal consistency. Cohort: 62 infants and their mothers were contacted in the first 72 hours after delivery through a Research Program from the Child Psychiatry Hospital Dr. Juan N. Navarro (Environmental Modulation of Infancy Development). They were born in two Mexico City hospitals: a state public hospital, Hospital de la Mujer (Women's Hospital) and in the Mother-infant Research Center from the Birth Study Group (CIMIGEN). All of them were low-risk infants who met the criteria for a one year follow up and whose parents gave their informed consent and accepted to be visited at their homes. The distribution of the infants group by sex was: 30 infants (48%) were female and 32 (52%) were male. Families were classified as follows: 72.5% were nuclear and 27.5% were extended. According to their socioeconomic profile, a high relationship was found between income and housing: seven (11.3%) of the highest income families lived in houses or apartments of their own or rented, where there were more rooms than inhabitants, whereas all others were living as follows: crowded houses or apartments (23 families, 37.1%), houses with of only one room (4 families, 6.5%) and the poorest houses with collective bathroom (28 families, 45.2%). All households had electrical lighting and most of them had also drinkable water inside their homes (98.4%); one shared water from a deep pond with their community. Bathrooms were: 35.5% with running water, 59.7% used a bucket for carrying water and 4.8% had latrine. Instrument: The Infant/Toddler HOME Inventory, version adapted by Cravioto, with 62 items in a binary format response was used. Subscale VII, Play Materials, asked about color, size, consistency or texture and type of toys. The instrument is administered by having a person calling the home at a time when the infant is awake and can be observed interacting with the mother or principal caregiver. The internal consistency of the Infant/Toddler HOME Inventory was monthly assessed for each of its subscales and the total scale, until the child's first year of life (12 months), applying Cronbach's alpha. Results showed that alpha coefficient's values higher than 0.60 throughout the 12 months were observed in the VII Play Materials shoed a range of 0.64 and 0.84, and 0.60 and 0.83 for the Total Scale. On the other hand, subscale VI, Physical Environment, showed a less than 0.60 value with an internal consistency coefficient of 0.56 in the first month of life; nevertheless the remaining months had values between 0.60 and 0.70. This was also observed with subscale II, Mental Development and Vocal Stimulation, with a 0.58 value for the first month of life and alpha values between 0.63 and 0.74 for the remaining eleven assessments. Subscale III, Emotional Climate, exhibited seven evaluations in a range between 0.61 and 0.76; this is the second with a 0.61, fourth with 0.69, sixth with 0.76, seventh with 0.67, ninth with 0.63, tenth with 0.69 and eleventh 0.63. Non-acceptable internal consistency, it is less than 0.60 in most of their scorings were observed for the following scales: I. Adult Contact, V. Breadth of Experience. Only in the sixth month's a value of 0.64 was observed, and the rest of scales exhibited a range within a 0.38 as inferior limit and 0.56 superior limit in the former and a 0.65 value in the tenth month and 0.60 in the eleventh. Subscale IV, Avoidance of Restriction, showed the lowest coefficient with values between 0.24 and 0.49. These results suggest that most of the subscales had adequate reliabilities, except for subscales I, Adult Contact; IV, Avoidance of Restriction; and V, Breadth of Experience. The purpose of this study was to evaluate the internal consistency of an instrument which measures the home environment quality in infants during their first year of life. Results showed the Infant/Toddler HOME Inventory had high internal consistency values in the Total Scale. These results are similar to those obtained by Banard, Bee & Hammond with a group of 179 children in Seattle, where they found 0.77 values at fourth months, 0.81 at eight and 0.86 at twelve. Our findings were within a rank of 0.68 and 0.83. Cronbach's alpha value obtained for the different subscales showed intersubscale differences. Subscales II, III, VI and VII showed internal consistency values equal or higher than 0.60 in most of the assessments during the 12 months of life. This implies measurements are precise and reliable when using them in low risk Mexican infants. Lower values found in subscales I, Adult Contact, IV, Avoidance of Restriction and V, Breadth of the Experience, may be due to the low number of items. According to Nunnally and Bernstein, it is recommended that measurements with a low internal consistency should be used with initial, non-crucial decisions, and with temporary and reversible effects susceptible to replication and rectification. It is very important to stress the fact that this Infant/Toddler HOME Inventory version is not comparable with other in the literature because the structure was notably modified. Limitations in this study are not only this modification, but the size sample, and future research efforts should overcome this fact by trying to sample infants from different Mexican regions. This study points out the need to adapt, develop and evaluate psychometrically instruments that measure specific aspects of the environment of infant's homes. Our results may be an initial step for those interested in measurements of Mexican families homes, or in those interested in the Infant/Toddler HOME Inventory as an indicator of the environmental aspects in early infancy.