Search

ABSTRACT Objectives. The study aims to assess the trend of neonatal, post-neonatal, and infant mortality from 1996 to 2020 within the metropolitan region of the state of Rio de Janeiro and other regions. Methods. Ecological study using the region as analysis unity. Data were accessed from the Mortality Information System and Live Birth Information System in the capital Rio de Janeiro, in the neighboring areas of Niterói, São Gonçalo, Baixada Fluminense, and the remaining regions of the state of Rio de Janeiro State. We applied Poisson multilevel modeling, where the models’ response variables were infant mortality and its neonatal and post neonatal components. Fixed effects of the adjusted models were region and death year variables. Results. During the 1996-2020 period, the Baixada Fluminense showed the highest infant mortality rate as to its neonatal and post neonatal components. All adjusted models showed that the more recent the year the lower the mortality risk. Niterói showed the lowest adjusted risk of infant mortality and its neonatal and post neonatal components. Conclusion. Baixada Fluminense showed the highest mortality risk for infant mortality and its neonatal and post-neonatal components in the metropolitan region. The stabilization in mortality rates in recent years was identified by the research.

Abstract: Although mortality from ischemic heart disease has declined over the past decades in Argentina, ischemic heart disease remains one of the most frequent causes of death. This study aimed to describe the role of individual and contextual factors on premature ischemic heart disease mortality and to analyze how educational differentials in premature ischemic heart disease mortality changed during economic fluctuations in two provinces of Argentina from 1990 to 2018. To test the relationship between individual (age, sex, and educational level) and contextual (urbanization, poverty, and macroeconomic variations) factors, a multilevel Poisson model was estimated. When controlling for the level of poverty at the departmental level, we observed inequalities in premature ischemic heart disease mortality according to the educational level of individuals, affecting population of low educational level. Moreover, economic expansion was related to an increase in ischemic heart disease mortality, however, expansion years were not associated with increasing educational inequalities in ischemic heart disease mortality. At the departmental level, we found no contextual association beween area-related socioeconomic level and the risk of ischemic heart disease mortality. Despite the continuing decline in ischemic heart disease mortality in Argentina, this study highlighted that social inequalities in mortality risk increased over time. Therefore, prevention policies should be more focused on populations of lower socioeconomic status in Argentina.

ABSTRACT Objectives: to investigate studies that adopted the multilevel analysis model to identify behavioral and structural risk factors associated with HIV infection. Methods: an integrative review of the literature with studies available in full, obtained from EMBASE, CINAHL, Pubmed, and Scopus, whose selected descriptors were the indexed terms: “HIV”, “multilevel analysis” and “behavior”. Results: the search resulted in 236 studies. Out of these, ten studies comprised the sample. Economic disadvantage, neighborhood characteristics, housing instability, incarceration, transactional sex, multiple partners, substance abuse, and age at first intercourse were classified as structural and behavioral risk factors for HIV. Reduced socioeconomic disadvantage, provision of housing stability, and condom use were associated with protective factors for HIV exposure. Conclusions: by applying the multilevel model in risk factor research studies, it was possible to identify the structural and behavioral elements of HIV risk.

Introduction: Self-rated health is strongly associated with morbidity and mortality. It is largely influenced by individual factors but also by individuals’ social surroundings and environment. Objective: To investigate individual, household, and locality factors associated with self-rated ealth in Colombian adults. Materials and methods: We conducted a cross-sectional multilevel study using data from national databases on 19 urban localities and 37,352 individuals nested within 15,788 households using a population-based survey. Given the natural hierarchical structure of the data, the estimates of self-rated health related to individual, household, and locality characteristics were obtained by fitting a three-level logistic regression. Results: The adjusted multilevel logistic models showed that at individual level, higher odds of poor self-rated health were found among older adults, persons from low socio-economic status, those living without a partner, with no regular physical activity, and reporting morbidities. At the household level, poor self-rated health was associated with households of low socioeconomic status located near noise sources and factories and in polluted and insecure areas. At the locality level, only poverty was associated with poor self-rated health after adjusting for individual and household variables. Conclusions: These results highlight the need for a more integrated framework when designing and implementing strategies and programs that aim to improve health conditions in urban populations in Latin America.

En este artículo se analiza una selección de cartas escritas por migrantes y de imágenes fotográficas realizadas a ellos y recogidas en la obra Enciclopedia de los Migrantes, fruto de una investigación y de la creación artística financiado por la UE y elaborado simultáneamente en 8 ciudades de Francia, España, Portugal y Gibraltar. Las cartas están dirigidas a una persona que quedó en su tierra de procedencia y explican sus vivencias personales, sus emociones, sentimientos y percepciones en relación al proceso migratorio. Las imágenes fueron realizadas por un fotógrafo artístico bajo la premisa de que tanto el escenario como los objetos retratados fueran elegidos por la propia persona migrante. Los retratos realizados fueron concebidos como una clase de etnofotografia, en la que se analiza tanto las imágenes y sus contenidos (mediante el método de análisis iconológico), particularmente la concepción del cuerpo y de las emociones de los migrantes. This article analyzes a selection of letters written by migrants and photographic images made to them, and collected in the work Encyclopedia of Migrants. The work is the result of research funded by the EU and developed, simultaneously, in 8 cities in France, Spain, Portugal and Gibraltar. The letters are addressed to a person who stayed in their homeland and explain their personal experiences, their emotions, feelings and perceptions regarding the migratory process. The images were made by an artistic photographer under the premise that both the stage and the objects portrayed were chosen by the migrant himself. The portraits were conceived as a kind of ethnophotography in which the images and their contents are analyzed simultaneously (through the method of iconological analysis), particularly the conception of the body and the emotions of the migrants.

Este artículo estudia por medio de los modelos lineales jerárquicos, la hipótesis de incidencia positiva que las variables de pedagogía y compromiso educativo generan sobre el desempeño académico. Para ello, son analizados las puntuaciones obtenidas por 62769 estudiantes de educación media en las pruebas PISA 2012 en 12 países. Al precisar que los procesos educativos adoptan una estructura jerárquica definida en dos niveles, estudiantes y planteles educativos, la especificación econométrica empleada conduce a resultados ostensiblemente mejores a los que se obtendría de una especificación tradicional. Entre los resultados se encuentra que el desempeño educativo en seis países es explicado en una mayor proporción por variables del entorno familiar y características del alumnado, mientras que en los seis restantes es explicado por el entorno escolar. De igual forma, los resultados validan la hipótesis de investigación formulada en el documento: la pedagogía y el compromiso educativo inciden positivamente sobre el desempeño académico. Así, la formulación de programas que fomenten dichas prácticas al interior de los planteles educativos debe formar parte de la agenda educativa. Por último, se encontró que el sexo del estudiante no aporta explicación alguna que refiera a existencia de brechas educativas pues los resultados son diferentes entre países, mientras que la no repetición de curso y la ausencia escolar, la dotación de elementos escolares en el hogar, el nivel de formación de los padres, entre otros factores, se asocian con sobresalientes resultados en el logro académico. || This article studies from the hierarchical linear models, the hypothesis of positive incidence that the variables of pedagogy and educational commitment generate on the academic performance. For this, the scores obtained by 62769 middle school students in the 2012 PISA tests in 12 countries are analyzed. By specifying that educational processes adopt a hierarchical structure defined in two levels: students and schools; the econometric specification used leads to results ostensibly better compared to those generated from a traditional specification. Among the results, we have that the educational performance in six countries is explained in a greater proportion by variables of the family environment and student’s characteristics, while in six it is explained by the school environment. Similarly, the results confirm the research hypothesis formulated in the document: pedagogy and educational commitment have a positive impact on academic performance, therefore, the formulation of programs that encourage such practices within the schools should be part of the educational policy. Finally, it was found that the sex of the student does not provide any explanation that refers to the existence of educational gaps because the results are dissimilar between countries, while the non-repetition and absence from school, the provision of school elements at home, the level of parent training, among other factors, are associated with outstanding results in academic achievement.

Students cannot change the composition and the size of the classroom they attend, and the effect of these variables on achievement is a source of educational inequality. In this study, based on a representative sample of Italian high school students (N = 26,799), a multilevel analysis was performed to: (i) estimate the overall impact of the structural characteristics of the class on achievements; (ii) identify which of the analysed characteristics were more relevant. The results show that the structure of the classroom has a significant impact on academic performance, and they identify group’s initial learning level and gender composition as the most important factors. The overall socio- economic level of the class and the percentage of immigrant students proved to have a lower impact, which varied according to different geographical areas, and the group size does not seem to be relevant. In conclusion, if the outcome of interest is academic performance, results suggest that the classroom size and the number of immigrant students attending the class has little or no relevance in high school, while the group’s initial achievement, the gender composition and, to a lesser extent, the socio-economic composition of the group should be taken into consideration. Given the complexity of the objectives of the educational system in Italy, which are by no means limited to the improvement of academic results, future studies will have to examine the effects of the structure of the classroom also on noncognitive outcomes that are closely related to students’ personal and social development.

Abstract: Few studies have investigated the simultaneous effect of individual and contextual factors on the occurrences of anemia. This study aims to evaluate the variability of children’s hemoglobin levels from municipalities in social vulnerability and its association with factors of individual and municipal nature. This is a cross-sectional, multi-center study, with children data (12-59 months) collected from 48 municipalities of the Southern region of Brazil, that were included in the Brazil Without Poverty Plan. Individuals’ data were collected using a structured questionnaire, and secondary and ecological data of children’s municipalities were collected via national surveys and institutional websites. The outcome was defined as the hemoglobin level obtained by HemoCue. A multilevel analysis was performed using Generalized Linear Models for Location Scale and Shape using R, with a 5% significance level. A total of 1,501 children were evaluated. The mean hemoglobin level was 12.8g/dL (95%CI: 12.7-12.8), with significant variability between municipalities. Lower values of hemoglobin were observed in children who lived in municipalities with a higher urbanization rate and a lower number of Community Health Agents, in relation to the reference categories. At the individual level, lower hemoglobin values were identified for children under 24 months, not enrolled at daycares, who were beneficiaries of the conditional cash transfer program and diagnosed with underweight. The results shed light on important factors at the municipal and the individual levels that were associated to the hemoglobin levels of children living in municipalities in social vulnerability.

Abstract: This cross-sectional study investigated the association between alcohol use by adolescents and the availability of alcohol outlets in the neighborhood of public schools. We collected primary data including variables at individual and school neighborhood level. Multilevel logistic regression was used to estimate odds ratio (OR) with a 95% confidence interval (95%CI) for alcohol use with the exposure variables. A total of 18.4% (95%CI: 13.2; 24.1) of adolescents reported using alcohol, which was associated with variables on the individual level such as being 18 years or older, working, and having previously smoked. Lower alcohol use was observed among adolescents from schools that were located 250m or more from alcohol outlets (OR = 0.29; 95%CI: 0.17; 0.48). Actions to reduce the use of alcohol among adolescents should take student’s and school neighborhood’s characteristics into account.

ABSTRACT We analyzed data from the National School-based Health Survey (PeNSE) carried out in Brazil in 2015 (n = 102,072 adolescents) to estimate how much of the individual variance in the prevalence of health behaviors is attributable to the school level. Multilevel logistic regression models were calculated to estimate the variance partitional coefficient (VPC) of the use of drugs, intake of unhealthy food, leisure physical activity and weight-related behaviors. The between-schools variance was significant in all tested models. The highest VPCs were observed when the use of drugs was analyzed (15%-20% of the total variance of smoking and use of illegal drugs). Lower, but still significant, values were observed in the other outcomes. The school context plays an important role in the adolescents’ health and should be considered in the design of public policies and actions in public health.

Abstract: This study aims to quantify the overall importance of schools in explaining the individual variance of tobacco use and to test the association between characteristics of the school environment and its vicinity with the experimentation and current use of cigarettes. We analyzed data from 102,072 Brazilian adolescents interviewed in the 2015 National School Health Survey (PeNSE). Multilevel logistic regression models were performed to estimate the between-schools variance and to test the association between school-level variables and the use of tobacco. Violence in the vicinity of the school and presence of teachers or students smoking on school premises were the school-level characteristics. The analyses were adjusted by individual covariates and stratified by gender. Around 12.5% of the individual variance in ever smoking was explained by between-school variation among girls (9.2% among boys). The figures were even higher for current smoking (14.9% girls; 12.2% boys) and current use of other tobacco products (27.7% girls; 17.8% boys). In general, the use of tobacco was associated with the existence of violence in the vicinity of the schools and was higher among students whose schools reported that students and teachers (teachers only for use of other tobacco products among girls) smoke on school premises. Tobacco use on school premises and the safety of the neighborhood where the school is located are associated with some smoking behaviors among adolescents. Such findings reinforce the necessity to effectively consider interventions in the school environment and neighborhood to fight smoking among adolescents.

Purpose: Test the influence of cultural orientation to innovation on the number of innovations measured in Brazilian companies. Originality/value: The study uses an approach that has not yet been explored in the area, the multilevel analysis. Design/methodology/approach: Five cultural orientation factors for innovation were considered: Rituals and Stories; Innovation strategies; Symbols and Heroes; Autonomy; and Standards and Communication. Innovation has been subdivided into four types: product, service, technology, and administration. The study consisted of a sample of 345 individuals in 60 organizations. The relationship between variables was analyzed using both linear regression and multilevel modeling. Findings: The results of the regression analysis showed positive and negative relationships of all cultural factors with some kind of innovation. The multilevel analysis indicated a relationship between the two variables only in the factors Rituals and Stories and Symbols and Heroes, both at the individual level. The study indicates the existence of organizational culture characteristics that favor and inhibit innovation. In addition, the complexity of the subject and the consequent need for further studies are highlighted. Objetivo: Testar a influência da orientação cultural à inovação no número de inovações mensuradas em empresas brasileiras. Originalidade/valor: O estudo utiliza uma abordagem que ainda não foi explorada na área, a análise multinível. Design/metodologia/abordagem: Cinco fatores de orientação cultural à inovação foram considerados: rituais e histórias; estratégias de inovação; símbolos e heróis; autonomia; e normas e comunicação. A inovação foi subdividida em quatro tipos: produto, serviço, tecnológico e administrativo. O estudo consistiu em uma amostra de 345 indivíduos em 60 organizações. A relação entre as variáveis foi analisada usando regressão linear e modelagem multinível. Resultados: Os resultados da análise de regressão mostraram relações positivas e negativas de todos os fatores culturais com algum tipo de inovação. A análise multinível indicou uma relação entre as duas variáveis apenas nos fatores rituais e histórias e símbolos e heróis, ambos no nível individual. O estudo indica a existência de características da cultura organizacional que favorecem e inibem a inovação. Além disso, destacam-se a complexidade do assunto e a consequente necessidade de novos estudos.

ABSTRACT OBJECTIVE To evaluate whether the scheduling model influences the perception of the user about the quality of primary health care centers. METHODS This is a cross-sectional and population-based study that measured the quality of centers by the Primary Care Assessment Tool (PCATool-Brazil), applied to adult users (n = 409) from 11 health centers in Florianópolis, state of Santa Catarina. Multilevel analysis was used to verify the relationship between the score of general quality of the primary health care and the scheduling model. The independent variables (age, skin color, scheduling model, panel size by primary health team, poverty ratio as income proxy, number of health teams, presence of economically interest areas, number of medical appointments in one year per primary health team, number of people treated in one year per health team), with p < 0.20 were selected for the multilevel model, which was adjusted with aggregates of information from users and health centers. RESULTS The health center that used advanced access had a general score of 7.04, while those using a weekly carve-out had a score of 6.26; the carve-out every 15 days, score of 5.87; and the traditional carve-out, score of 6.29. CONCLUSIONS The scheduling model of advanced access had a positive effect on the quality of primary health care, in the perception of users.

This topic review aims to present a global vision of multilevel analysis models’ applicability to health research, explaining its theoretical, methodological, and statistical foundations. We describe the basic steps to build these models and examples of their application according to the data hierarchical structure. It ir worth noticing that before using these models, researchers must have a rationale for needing them, and a statistical evaluation accounting for the variance percentage explained by the observations grouping effect. The requirements to conduct this type of analysis depends on special conditions such as the type of variables, the number of units per level, or the type of hierarchical structure. We conclude that multilevel analysis models are a useful tool to integrate information, considering the complexity of the relationships and interactions involved in most health conditions, including the loss of independence between observation units.

Abstract: This study aimed to identify the main regional factors associated with variations in the prevalence of functional limitation on the older adult in Colombia adjusted by individual characteristics. This multilevel study used cross-sectional data from 23,694 adults over 60 years of age in the SABE, Colombia nationwide survey. State-level factors (poverty, development, inequity, violence, health coverage, and access to improved water sources), as well as individual health related, socioeconomic and demographic characteristics, were analyzed. The overall prevalence of functional impairment for the basic activities of daily living (ADL) was 22%. The presence of comorbidities, low educational level, physical inactivity, no participation in social groups, mistreatment and being over 75 years old were associated with functional limitation. At the group level, the analysis showed significant differences in the functional limitation prevalence across states, particularly regarding the socioeconomic status measured according to the Human Development Index (median OR = 1.22; 95%CI: 1.13-1.30; p = 0.011). This study provides evidence on the impact of socioeconomic variation across states on FL prevalence in the Colombian elderly once adjusted for individual characteristics. The findings of this study, through a multilevel approach methodology, provide information to effectively address the conditions that affect the functionality in this population through the identification and prioritization of public health care in groups with economic and health vulnerability.

The objective of this research was to examine the association between income inequality and high blood pressure in Colombia. Using a nationally representative Colombian sample of adults, and data from departments and municipalities, we fit sex-stratified linear and logistic multilevel models with blood pressure as a continuous and binary variable, respectively. In adjusted models, women living in departments with the highest quintile of income inequality in 1997 had higher systolic blood pressure than their counterparts living in the lowest quintile of income inequality (mean difference 4.42mmHg; 95%CI: 1.46, 7.39). Women living in departments that were at the fourth and fifth quintile of income inequality in 1994 were more likely to have hypertension than those living in departments at the first quintile in the same year (OR: 1.56 and 1.48, respectively). For men, no associations of income inequality with either systolic blood pressure or hypertension were observed. Our findings are consistent with the hypothesis that income inequality is associated with increased risk of high blood pressure for women. Future studies to analyze pathways linking income inequality to high blood pressure in Colombia are needed.

Objective: To examine variation in the duration of non-work-related sickness absence (NWRSA) across geographical areas and the degree to which this variation can be explained by individual and/or contextual factors. Methods: All first NWRSA episodes ending in 2007 and 2010 were analyzed. Individual (diagnosis, age, sex) and contextual factors (healthcare resources, socioeconomic factors) were analyzed to assess how much of the geographical variation was explained by these factors. Median NWRSA durations in quartiles were mapped by counties in Catalonia. Multilevel Cox proportional hazard regression models with episodes nested within counties were fitted to quantify the magnitude of this variation. The proportional change in variance (PCV), median hazard ratios (MHR) and interquartile hazard ratios (IHR) were calculated. Results: We found a geographical pattern in the duration of NWRSA, with longer duration in northwestern Catalonia. There was a small, but statistically significant, geographical variation in the duration of NWRSA, which mostly decreased after adjustment for individual factors in both women (PCV = 34.98%, MHR = 1.09, IHR = 1.13 in 2007; PCV = 34.68%, MHR = 1.11, IHR = 1.28 in 2010) and men (PCV = 39.88%, MHR = 1.10, IHR = 1.27 in 2007; PCV = 45.93%, MHR = 1.10, IHR = 1.25 in 2010); only in the case of women in 2010 was there a reduction in county-level variance due to contextual covariates (PCV = 16.18%, MHR = 1.12, IHR = 1.32). Conclusions: County-level variation in the duration of NWRSA was small and was explained more by individual than by contextual variables. Knowledge of geographic differences in NWRSA duration is needed to plan specific programs and interventions to minimize these differences.

Objective. To investigate prevalence of poor self-rated health and its association with individual and household-level characteristics among adults and elderly in Brazil. Materials and methods. Cross-sectional study with Brazilian National Household Sample Survey 2008 (n=257 816). Crude and multilevel-adjusted Poisson regression models were fitted. Results. After adjusted analysis, poor self-rated health was significantly associated with higher household income, living alone, not having piped water nor garbage collection, lower education, not having health insurance, female sex, higher age, being a current or previous smoker, physical inactivity, having chronic diseases, having physical impairment. Subjects living in rural areas also had higher prevalence of poor self-rated health. The factors most strongly associated with the outcome were physical impairment and reporting three or more chronic diseases. Conclusions. Socioeconomic, health related behaviors, and physical health were associated with poor self-rated health.

This study was focused on parents’ socialization values, namely the values that parents want their children to adopt, and their sources. In a sample of 325 Italian families with one adolescent child (14-18 years), it aimed at comparing fathers’ and mothers’ socialization values and assessing parents’ own personal values and family value climate as antecedents of the values parents would like their children to endorse. For each family both parents and the adolescent were involved and asked to complete the Portrait Values Questionnaire individually. The Anova results showed significant differences between fathers’ and mothers’ socialization values: in particular, fathers gave more importance to openness to change and self-enhancement values in their children’s rearing than mothers did. Using multilevel analysis, as fathers and mothers were nested within families, we found significant and positive relations between parents’ personal values and all their socialization values, as well as between family value climate and some of the parents’ socialization values. Conversely, cross-level interactions between parents’ personal values and family value climate did not contribute to predict the values parents want their children to adopt. Implications of this research and its possible developments are discussed.

Objective: To determine the prevalence of negative mood states in adolescents according to gender, to analyze variability among schools, and to evaluate the associated factors. Methods: A cross-sectional study with a cluster design was carried out. We administered the High-school students health survey to a sample of 9,340 students (aged 14-16 years) in the third and fourth year of Compulsory Secondary Education in Catalonia, Spain, during the 2005-6 academic year. The main outcome measure was evidence of a negative mood state. A multilevel logistic regression model stratified by gender was used to identify the factors associated with negative mood states and to determine variability among distinct schools. Results: Approximately 19% of adolescents reported evidence of a negative mood state, with a higher prevalence in girls (25%). The most significant factors associated with negative mood states were "use of tranquilizers" and "having eating disorders" in girls and "not exercising" and "poor self-perception of health status" in boys. In both genders, variability was found among schools in the prevalence of negative mood states (girls: variance = 0.078; p Objetivos: Determinar la prevalencia del estado de ánimo negativo entre alumnos adolescentes según sexo, analizar la variabilidad entre escuelas y evaluar los factores asociados. Métodos: Estudio transversal basado en un muestreo por conglomerados bietápico. Administramos una encuesta de salud a 9340 estudiantes de tercero y cuarto curso de Educación Secundaria Obligatoria de 14 a 16 años de edad, en Cataluña, durante el curso escolar 2005-06. La variable principal fue el estado de ánimo negativo. Se usó un modelo de regresión logística multinivel estratificado por sexo para identificar los factores asociados al estado de ánimo negativo y determinar la variabilidad entre las diferentes escuelas. Resultados: Aproximadamente el 19% de los adolescentes refirieron un estado de ánimo negativo, siendo más prevalente entre las chicas (25%). En ellas, los factores asociados significativamente con estados de ánimo negativo fueron «tomar tranquilizantes» y tener trastornos alimentarios, mientras que en los chicos fueron no realizar deporte y tener una mala percepción de su estado de salud. Hay variabilidad en la prevalencia del estado de ánimo negativo según las escuelas en ambos sexos (chicas: varianza = 0,078, p

No presente estudo objetivou-se identificar quais características de alunos e escolas são associadas ao desempenho em Português, bem como verificar se existem diferenças entre as regiões geográficas brasileiras. Foram analisados os dados de 33.962 alunos da 3ª série do Ensino Médio, distribuídos em 1.661 escolas brasileiras pesquisadas no SAEB/2001. Entre as variáveis de controle, a maior contribuição para o modelo final foi da variável nível socioeconômico da escola. Pelo menos 41% da variância do nível da escola e 14% da variância do nível do aluno foram explicados no modelo multinível final. Concluí-se que a desigualdade entre regiões do Brasil em relação ao desempenho médio das escolas em Língua Portuguesa, é fortemente relacionada com o nível socioeconômico dos alunos e das escolas.El presente estudio tiene por objeto identificar qué características de los alumnos y de las escuelas se asocian al rendimiento en Lengua Portuguesa, y verificar si existen diferencias entre las regiones geográficas brasileñas. Para tal efecto se analizaron los datos de 33.962 alumnos del 3º año de la enseñanza secundaria, distribuidos en 1.661 escuelas brasileñas analizadas en la evaluación de 2001 hecha por el SAEB (Sistema de Evaluación de la Educación Básica). Entre las variables de control, la mayor contribución para el modelo final fue la variable nivel socioeconómico de la escuela. Por lo menos el 41% de la varianza del nivel de la escuela y el 14% de la varianza del nivel de alumno se explicaron en el modelo multinivel final. Así se concluye que la desigualdad entre las regiones de Brasil, referente al rendimiento medio de las escuelas en Lengua Portuguesa, se relaciona bastante con el nivel socioeconómico de los alumnos y de las escuelas.The purpose of the present study was to identify which characteristics of students and schools affect proficiency in Portuguese Language, as well as to verify the existence of differences between Brazilian regions in this respect. The data were analyzed of 33,962 third grade students nessed in 1,661 secondary schools who participated in the 2001 assessment of the SAEB. Among the control variables, socioeconomic status of the school had the strongest contribution to the final model. At least 41% of the school level variance and 14% of the student level variance were explained in the final multilevel model. It can be concluded that the inequality between Brazilian regions in relation to the performance of schools in Portuguese Language is strongly related to the socioeconomic status of the students of the schools.

Competitive advantage is the main hypothesis to explain the firm performance heterogeneity. However, competitive advantage is frequently simplified and empirically treated as profitability, despite its implications for the overall firm performance. Considering that, this article develops a metric for competitive advantage by means of the analysis of its effects over financial performance, combining the curves of profitability and market share growth. Using a multilevel model, the individual results for each firm is isolated and tested against each sector mean. The model is applied to a sample of American companies (COMPUSTAT), in four intervals during the period of 1990 to 2009. The results reveal different competitive configurations, in which 16% of the firms are positioned in advantage and 16,5% in disadvantage (2005-2009),considering that the analysis for the other intervals demonstrated stability of the results.

This paper investigates whether there are different patterns of traditionality in different countries with regard to a perceived negative impact of labor-force participation of mothers on their children and family life. For this purpose, individual-level traditionality subgroups and segments of countries with different traditionality patterns of their nationals were identified simultaneously by means of multilevel latent-class (ML-LC) analysis of the answers to three items of the Changing Family and Gender Roles module of the International Social Survey Program (ISSP). This module was fielded in 22 countries in the years 1994 and 2002. Six individual-level subgroups and five country segments can be discerned. The structure of individual-level subgroups is almost identical in both years. Four individual-level subgroups differ only quantitatively in their level of traditionality. Two further subgroups are characterized by a unique tendency to defend working mothers against criticism. From 1994 to 2002 the sizes of traditional subgroups decrease, and there is also some change in the composition of country segments. This paper investigates whether there are different patterns of traditionality in different countries with regard to a perceived negative impact of labor-force participation of mothers on their children and family life. For this purpose, individual-level traditionality subgroups and segments of countries with different traditionality patterns of their nationals were identified simultaneously by means of multilevel latent-class (ML-LC) analysis of the answers to three items of the Changing Family and Gender Roles module of the International Social Survey Program (ISSP). This module was fielded in 22 countries in the years 1994 and 2002. Six individual-level subgroups and five country segments can be discerned. The structure of individual-level subgroups is almost identical in both years. Four individual-level subgroups differ only quantitatively in their level of traditionality. Two further subgroups are characterized by a unique tendency to defend working mothers against criticism. From 1994 to 2002 the sizes of traditional subgroups decrease, and there is also some change in the composition of country segments.Este trabajo analiza si existen diferentes patrones de tradicionalismo en los diferentes países sobre la percepción de un impacto negativo de la participación laboral de las madres en sus hijos y en su vida familiar. Para ello se identificaron, de forma simultánea, subgrupos de tradicionalismo a nivel individual y segmentos de países con diferentes patrones de tradicionalismo de sus ciudadanos, a través de un análisis multinivel de clases latentes (ML-LC) de las respuestas a tres preguntas del módulo Familia y Roles de Género del International Social Survey Program (ISSP), que se aplicó en 22 países en los años 1994 y 2002. Se identificaron seis subgrupos de nivel individual y cinco segmentos de países. La estructura de los subgrupos a nivel individual es casi idéntica en ambos años. Cuatro subgrupos a nivel individual difieren solamente de manera cuantitativa en su nivel de tradicionalismo. Dos subgrupos más se caracterizan por una tendencia a defender de la crítica a las madres que trabajan. De 1994 a 2002 el tamaño de los subgrupos tradicionales disminuye, y también se ha producido algún cambio en la composición de los segmentos de países.

Objetivos: Analizar variables asociadas al uso de los servicios de salud bucodental en el último año por parte de la población de 6 a 15 años de edad residente en España. Analizar si hay variabilidad en el uso de estos servicios entre comunidades autónomas y si ésta podría explicarse por variables relacionadas con el modelo asistencial. Métodos: Estudio transversal de la Encuesta Nacional de Salud (2006). Variables independientes de índole individual (sociodemográficas, enfermedad, hábitos y socioeconómicas) y contextual (modelo asistencial y prevalencia de paro en la comunidad autónoma). Estimación de la asociación mediante regresión logística multinivel. Resultados: La varianza en el uso entre comunidades autónomas es de 0,16 (EE: 0,07) y el 4,8% de la variabilidad total es atribuible a la comunidad autónoma. Las variables incluidas en el modelo explican el 83,11% de la varianza existente entre comunidades autónomas. Las variables individuales asociadas a una mayor probabilidad de uso fueron la presencia de enfermedad y la frecuencia de cepillado de los dientes; las asociadas a una menor probabilidad fueron la edad, la procedencia, la ingesta de refrescos azucarados y el nivel socioeconómico. Entre las variables contextuales, estar cubierto por un modelo asistencial tipo Plan de Asistencia Dental Infantil (PADI), implantado desde hace más de 10 años, duplica la probabilidad de acudir al dentista frente a los que carecen de él (odds ratio=2,47; intervalo de confianza del 95%: 2,04-2,99). Conclusiones: El uso en el último año de los servicios de salud bucodental por la población infanto-juvenil residente en España se aleja de las recomendaciones. Se identifican variables individuales (sociodemográficas, de enfermedad, de hábitos y socioeconómicas) y contextuales (modelo asistencial) asociadas a dicho uso.Objectives: To analyze the variables associated with the use of oral health services in the last year by the population aged 6 to 15 years living in Spain and to determine whether there is variability in the use of these services among autonomous regions and, if so, whether this variability could be explained by variables related to the care model of the distinct autonomous regions. Methods: A cross-sectional study of the Spanish National Health Survey (2006) was carried out. Independent variables were individual (sociodemographic, dental disease, habits and socioeconomic) and contextual (type of dental care model and prevalence of unemployment in the autonomous region). Association was estimated by multilevel logistic regression. Results: Variance in the use of oral health services among autonomous regions was 0.16 (SE: 0.07), and 4.8% of the total variability was attributable to the autonomous region. The variables included in the model explained 83.11% of the variance. Individual variables associated with an increased likelihood of using dental services were the presence of disease and the frequency of brushing. Individual variables associated with a lower likelihood were age, origin, intake of sugary soft drinks and socioeconomic status. The contextual variables of being covered by a dental care model (of the type Childhood Dental Care Plan) older than 10 years doubled the likelihood of using oral health services compared with those without such coverage (OR=2.47, CI=2.04-2.99). Conclusions: The use of oral health services during the last year by the pediatric population in Spain is lower than recommended. This use is associated with individual variables (demographic, dental health, habits and socioeconomic factors) and contextual variables (dental care model).

Using a representative survey of Chilean journalists (n= 570) and data collected from each news media organization (n= 114), this paper examines individual and organizational effects on professional role conceptions via hierarchical linear modeling. The findings demonstrate the existence of significant contextual effects on the importance that journalists give to their different functions in society, although not all professional roles are equally affected in form and magnitude by the news media context. This study shows that the journalism’s more political roles are modeled for an interrelation between individual and organizational factors, where journalists’ background and beliefs tend to be more predictive than news media characteristics. In turn, professional functions linked to the need of the public and to the commercialization of news, appeared to be more tied to the structural characteristics of the news organizations where the journalists belong. The results also reveal that individual and organizational political factors predict the majority of the professional roles, while ownership and media type do not have any direct influence on the journalists’ attitudes.

O objetivo deste estudo ecológico longitudinal foi analisar a tendência da proporção de baixo peso ao nascer no Rio Grande do Sul, Brasil, de 1994 a 2004, utilizando a análise de dados de painel e regressão linear multinível (dois níveis: microrregião e tempo (anos)) para estimar os fatores de risco associados à proporção de baixo peso ao nascer. A proporção de baixo peso ao nascer teve um crescimento anual de 1,2%, e o modelo multinível mostrou que as proporções diferem entre as microrregiões e aumentam em associação com os anos, com o aumento do percentual de prematuros, com o aumento do coeficiente de mortalidade infantil e com o aumento do percentual de cesarianas. Entre as microrregiões, as proporções de baixo peso ao nascer variam positivamente com o percentual de urbanização, com os gastos com o Sistema Único de Saúde e negativamente com o percentual de participação na atividade econômica. O modelo multinível mostrou que a maior parte da variação nas proporções de baixo peso ao nascer se deve aos efeitos da microrregião de moradia da mãe do nascido vivo.The aim of this longitudinal ecological study was to analyze the trend in the proportion of low birth weight in Rio Grande do Sul State, Brazil, from 1994 to 2004 by panel data analysis and multilevel linear regression (two levels: by micro-region and time in years) to estimate risk factors associated with low birth weight. The proportion of low birth weight increased by 1.2% per year, and the multilevel model showed that the proportions differed between the micro-regions and increased over time, with the increase in the percentage of premature newborns, with the increase in the infant mortality rate, and with the increase in the cesarean rate. Among the micro-regions, the proportions of low birth weight varied positively with the urbanization rate and expenditures in the Unified National Health System and negatively with rate of participation in the workforce. According to the multilevel model, most of the variation in proportions of low birth weight was due to the effects of the micro-region of residence of the newborn's mother.

OBJETIVO: Avaliar a efetividade de programa governamental de suplementação alimentar no ganho ponderal de crianças. MÉTODOS: Estudo de coorte com dados secundários de 25.433 crianças de baixa renda com idade entre seis e 24 meses que ingressaram em programa de distribuição de leite fortificado Projeto Vivaleite, realizado no Estado de São Paulo de 2003 a 2008. O ganho ponderal foi medido por meio dos valores de escores z de peso para idade, calculados pelo padrão da Organização Mundial da Saúde (2007), obtidos, na rotina do programa, ao ingressar e a cada quatro meses durante a permanência. As crianças foram divididas em três grupos de escore z ao entrar: sem comprometimento de peso (z > -1); risco de baixo peso (-2 < z < -1) e baixo peso (z < -2). Utilizou-se regressão linear multinível (modelo misto), permitindo a comparação, em cada idade, das médias ajustadas do escore z dos ingressantes e participantes há pelo menos quatro meses, ajustadas para correlação entre medidas repetidas. RESULTADOS: Verificou-se efeito positivo do programa no ganho de peso das crianças, variando em função do estado nutricional ao ingressar; para as que entraram sem comprometimento de peso, o ganho médio ajustado foi 0,183 escore z;entre as que entraram com risco de baixo peso, foi 0,566; e entre as ingressantes com baixo peso, foi 1,005 escore z. CONCLUSÕES: O programa é efetivo para o ganho ponderal de crianças menores de dois anos, com efeito mais pronunciado entre as crianças que entram no programa em condições menos favoráveis de peso.OBJETIVO: Evaluar la efectividad del programa gubernamental de suplementación alimentaria en la ganancia ponderal de niños. MÉTODOS: Estudio de cohorte con datos secundarios de 25.433 niños de baja renta con edad entre seis a 24 meses, que ingresaron en programa de distribución de leche fortificado Proyecto “Viva Leche” realizado en el Estado de Sao Paulo (sureste de Brasil) de 2003 a 2008. La ganancia ponderal fue medida por medio de los valores de escores z de peso por edad, calculados por el patrón de la Organización Mundial de la Salud (2007), obtenidos, en la rutina del programa, al ingresar y a cada cuatro meses durante la permanencia. Los niños fueron divididos en tres grupos de escore z al entrar: sin comprometer el peso (z>-1); riesgo de bajo peso (-2£zOBJECTIVE: To evaluate the effectiveness of a governmental supplementary feeding program in child weight gain. METHODS: A cohort study including secondary data on 25,433 low-income children aged between six and 24 months, participating in a fortified milk distribution program known as Projeto Vivaleite, was performed in the state of São Paulo, Southeastern Brazil, between 2003 and 2008. Weight gain was measured using weight-for-age z-score values, calculated according to the World Health Organization standards (2007). These values were obtained in the program routine, when a child started it and every four months during their stay. Children were divided into three z-score groups when starting the program: weight gain not compromised (z>-1); risk of low weight (-2

Objetivos: Tradicionalmente las variaciones en el consumo de bebidas alcohólicas se han estudiado en términos de las características individuales. Con este estudio nos propusimos valorar la contribución de los factores contextuales en conjunto con los individuales, en la cantidad de bebidas alcohólicas que los individuos consumen. Métodos: Se midieron variables en 2 niveles: individual y contextuales (vecindarios). Las primeras se obtuvieron de una muestra representativa de la población mayor de 15 años del municipio 10 de Octubre y las contextuales mediante entrevistas con los representantes del gobierno de las áreas del municipio. Se usaron 3 modelos multiniveles: modelo nulo, modelo de interceptos aleatorios y modelo de coeficientes dependientes. Resultados: Existen diferencias regionales en el consumo de bebidas alcohólicas y estas diferencias persisten después de ajustar por las características individuales, por lo que se le pueden atribuir a los contextos. De los factores contextuales estudiados, 4 tienen influencia directa en la cantidad de bebidas alcohólicas que los individuos consumen y 2 de ellos interactúan con la situación laboral de las personas. Por ejemplo, las personas que están desocupadas consumen más que las que tienen alguna ocupación, pero los desocupados que viven en vecindarios con numerosos puntos de venta de bebidas alcohólicas, consumen más que los desocupados que viven en vecindarios con menos puntos de venta. Conclusiones: Los modelos multiniveles son una herramienta importante que permiten estudiar los efectos individuales y contextuales que se relacionan con el consumo de bebidas alcohólicas. Los resultados de la aplicación de una estrategia de modelación multinivel sugieren que los factores individuales y contextuales desempeñan un papel importante en la estructuración del consumo de bebidas alcohólicas.Objectives: Traditionally the variations in alcohol consumption have been studied in terms of individual characteristics. In the present study we intended to assess the contribution of contextual factors and individual factors as well, in the units of alcohol consumed per person per. Methods: We measured variables in two levels: individual and contextual (neighborhoods). The first ones were obtained by means of a cross-sectional survey of people 15 years an older in Havana City. Contextual variables were obtained through interviews with the representatives of the local government in the city. Three multilevel models were used: null model, random intercept models and dependent coefficients model. Results: Our findings suggest that there are variations in the average number of units consumed among neighborhoods and these differences persist after adjusting for individual characteristics; so part of the variations can be attributed to the contexts. Six contextual factors were studied. Four have direct influence in the average number of units consumed and two of them interact with the economic situation of the respondents. For example, people that are unoccupied consume more than those that have an occupation; but the unoccupied ones that live in neighborhoods with numerous alcohol outlets consume more than the unoccupied ones that live in neighborhoods with less alcohol outlets. Conclusions: The results suggest that individual and contextual factors play an important role in structuring the patterns of alcohol consumption in order to design better health promotions and prevention strategies. The high-level alcohol drinkers tend to group in neighborhoods with a large number of alcohol outlets.

This study identified the agglomeration effect of firms classified in the same economic activity on the rate of growth of local employment. Data on all industrial firms located in State of São Paulo and included in the Government’s Annual List of Social Information over a ten-year period (1996-2005) were collected. Some 263,020 observations on the number of jobs corresponding to 26,231 combinations of municipalities and 296 economic activities were analyzed. To identify the local economic agglomerations, two sets of criteria (Puga, 2003; Suzigan, Furtado, Garcia, & Sampaio, 2003) were used. A growth curve model, using a multilevel model was developed with the Hierarchical Linear Models [HLM] software. The results showed a positive relationship between agglomeration of firms in the same economic activity and employment growth. Considering the expected positive externalities due to agglomeration of similar firms in the same municipality, the results suggest that firms located in an agglomeration can have higher growth rates than their counterparts located outside the agglomeration. The results are interesting and relevant to the individual firm but also to the support of public policy regarding regional development at the municipal level. The findings support previous case studies, giving robustness to theory.

OBJETIVO: Analisar a influência dos fatores demográficos, socioeconômicos, de condições de saúde e do contexto das unidades da federação na incapacidade funcional dos idosos. MÉTODOS: Estudo transversal que utilizou dados da Pesquisa Nacional por Amostra de Domicílio (PNAD) de 2003, do Instituto Brasileiro de Geografia e Estatística (IBGE) e do Instituto de Pesquisa Econômica Aplicada (Ipea). A amostra foi constituída de 33.515 indivíduos com 60 anos ou mais de idade. A variável dependente foi a incapacidade funcional, mensurada pela dificuldade por subir ladeira ou escada. As variáveis independentes foram divididas em dois níveis: individual (características demográficas, socioeconômicas e relativas à saúde) e de contexto (Índice de Gini e Produto Interno Bruto per capita por unidade da federação em 2000). Um modelo de regressão logística multinomial multinível foi utilizado para estimar o efeito das variáveis independentes na incapacidade funcional dos idosos. RESULTADOS: A incapacidade funcional foi associada com fatores demográficos, socioeconômicos e de saúde. Em nível individual, o sexo, a educação, a renda, a ocupação, a autopercepção de saúde e as doenças crônicas foram os fatores mais fortemente relacionados. Em nível de contexto, a desigualdade de renda exibiu uma importante influência. CONCLUSÕES: A autopercepção de saúde é o fator mais fortemente relacionado com a incapacidade funcional dos idosos no Brasil, seguida das doenças crônicas. Sexo, ocupação, escolaridade e renda também são altamente associados. Ações que abordam os principais fatores associados à incapacidade funcional podem contribuir significativamente para o bem-estar e qualidade de vida dos idosos.OBJETIVO: Analizar la influencia de los factores demográficos, socioeconómicos, de condiciones de salud y del contexto de las unidades de la federación en la incapacidad funcional de los ancianos. MÉTODOS: Estudio transversal que utilizó datos de la Pesquisa Nacional por Muestra de Domicilio (PNAD) de 2003, del Instituto Brasilero de Geografía y Estadística y del Instituto de Pesquisa Económica Aplicada. La muestra fue constituida de 33.515 individuos con 60 años o más de edad. La variable dependiente fue la incapacidad funcional, medida por la dificultad para subir ladera o escalera. Las variables independientes fueron divididas en dos niveles: individual (características demográficas, socioeconómicas y relativas a la salud) y de contexto (Índice de Gini y Producto Interno Bruto per capita por unidad federación en 2000). Un modelo de regresión logística multinomial multinivel fue utilizado para estimar el efecto de las variables independientes en la incapacidad funcional de los ancianos. RESULTADOS: La incapacidad funcional fue asociada con factores demográficos, socioeconómicos y de salud. En nivel individual, el sexo, la educación, la renta, la ocupación la autopercepción de salud y las enfermedades crónicas fueron los factores más fuertemente relacionados. En nivel de contexto, la desigualdad de renta exhibió una importante influencia. CONCLUSIONES: La autopercepción de salud es el factor más fuertemente relacionado con la incapacidad funcional de los ancianos en Brasil, seguido por las enfermedades crónicas. Sexo, ocupación, escolaridad y renta también son altamente asociados. Acciones que abordan los principales factores asociados a la incapacidad funcional pueden contribuir significativamente para el bienestar y calidad de vida de los ancianos.OBJECTIVE: To analyze the influence of demographic, socioeconomic, health, and contextual factors of the Brazilian federation units on the functional disability of the elderly. METHODS: Cross-sectional study based on data from the 2003 PNAD (Brazilian National Household Survey), of IBGE (Brazilian Institute of Geography and Statistics) and Ipea (Institute of Applied Economic Research). The sample was composed of 33,515 individuals aged 60 years and older. The dependent variable was functional disability, measured by difficulty in climbing slopes or stairs. The independent variables were divided into two levels: individual (demographic, socioeconomic and health-related characteristics) and contextual (Gini Index and Gross Domestic Product per capita by Brazilian state in 2000). A multinomial and multilevel logistic regression model was utilized in order to estimate the effect of the independent variables on the functional disability of the elderly. RESULTS: Functional disability was associated with demographic, socioeconomic and health factors. At the individual level, sex, level of schooling, income, occupation, self-perception of health and chronic diseases were the factors that were most strongly related to functional disability. At the contextual level, income inequality proved to exert an important influence. CONCLUSIONS: Self-perception of health is the factor that is most strongly related to the functional disability of the elderly in Brazil, followed by chronic diseases. Sex, occupation, level of schooling and income are also highly associated with it. Actions that approach the main factors associated with functional disability can contribute significantly to the well-being and quality of life of the elderly.

Objetivo: Estimar la asociación entre el bienestar material de la provincia de residencia y la inactividad física durante el tiempo libre en la población española mayor de 15 años. Métodos: Los datos proceden de la encuesta de Discapacidades, Deficiencias y Estado de Salud, realizada por el Instituto Nacional de Estadística en 1999. Se analizaron los datos de 24.561 mujeres y 21.133 varones. Se consideró un sujeto inactivo cuando declaraba no realizar ningún tipo de actividad física durante el tiempo libre. La medida de bienestar material fue la renta per cápita de la provincia agrupada en cuartiles. La medida de asociación entre la renta per cápita y la inactividad física fue la odds ratio (OR) estimada mediante modelos logit multinivel. Resultados: En los individuos ≥ 45 años de edad se encontró una asociación entre la renta per cápita y la inactividad física. Las provincias con mayor renta per cápita presentaron la menor prevalencia de inactividad física. No obstante, la OR de mayor magnitud se encontró en el cuartil 2 de renta per cápita, no observándose diferencias estadísticamente significativas entre el cuartil más rico y el cuartil más pobre. Al ajustar por características socioeconómicas y otras variables individuales, esa asociación persistió en las mujeres y desapareció en los varones. En las mujeres la asociación fue mayor en el grupo de ingresos personales más bajos. Conclusiones: Los resultados sugieren que en las mujeres mayores de 45 años la inactividad física podría estar relacionada no sólo con las características individuales, sino también con el contexto socioeconómico del área de residencia.Objective: To estimate the association between material wellbeing of the province of residence and leisure-time physical inactivity in the Spanish population aged 16 years and older. Methods: We used data from the Survey on Disabilities, Impairments and Health Status carried out by the Statistical National Institute in 1999. We analyse 24,561 women and 21,133 men. Respondents were classified as inactive if they reported no leisure-time physical activity. The measure of material wellbeing was the per capita income of the province of residence grouped in quartiles. The measure of the association between per capita income and physical inactivity was the odds ratio (OR) estimated from logit multilevel models. Results: Association between per capita income and physical inactivity was observed in people aged 45 years and older. The lower prevalence of physical inactivity was observed in provinces with the higher per capita income. However the higher OR was found in quartile 2 of per capita income, because no significant difference was observed between quartile 4 (richest) and quartile 1 (poorest). After adjusting for socioeconomic characteristics and other individual variables, association remained significant in women and disappeared in men. In women, association between per capita income and physical inactivity was higher in the population group with lower personal income. Conclusions: Results suggest that physical inactivity in women aged 45 years and older could be related not only with individual characteristics but with the socioeconomic context of the area of residence.

Objetivo: El propósito de este artículo es analizar el uso de la metodología multinivel en las ciencias de la salud. Diseño: Se llevó a cabo una búsqueda bibliográfica en Medline de los artículos publicados entre 1995 y 2001, utilizando 15 palabras de búsqueda, entre las que se encuentran «multilevel model», «multilevel analysis», «hierarchical linear model», «mixed model» y «random effects model», obteniéndose un total de 1.288 resúmenes de artículos. Se examinaron los 222 artículos teóricos y/o aplicaciones realmente relacionadas con métodos multinivel, para determinar el nivel de uso actual, el tipo de modelo, el número de niveles, las unidades de análisis y las variables de respuesta a través de las especialidades en salud. Resultados: El 66,7% de los artículos estudiados se encontraron con las palabras de búsqueda «multilevel analysis», «multilevel modelling» y «multilevel model». El 56% de los artículos son aplicaciones de multinivel en diferentes áreas de la salud; no obstante, merece destacar que un 10% de los trabajos tienen como objetivo exponer y difundir la metodología en un lenguaje comprensible para los no especialistas. Conclusiones: El uso de los modelos multinivel en salud ha aumentado con los años, aplicándose en diferentes áreas o especialidades de salud, pero aún no se los considera una técnica de uso habitual.Objective: The aim of this study was to analyze the use of multilevel methodology in health sciences. Design: A literature search was performed in Medline for articles published between 1995 and 2001. Fifteen search words were used, some of which were «multilevel model», «multilevel analysis», «hierarchical linear model», «mixed model», and «random effects model». A total of 1288 abstracts were retrieved. Two hundred twenty-two of the articles on theoretical and/or applied issues related to multilevel methods were examined to determine their current use, the type of model, the number of levels, the units of analysis and the outcome variables related to health sciences. Results: 66.7% of the articles studied was found with the keywords: «multilevel analysis», «multilevel modelling» and «multilevel model». Fifty-six percent of the articles were multilevel applications in different health sciences. However, in 10% of the articles, the main objective was to present and disseminate the methodology in a language comprehensible to non-specialists. Conclusions: The use of multilevel modelling in health has increased with time and is being applied in different health areas and specialties. However, it is still not considered a commonly used technique.

The present work is a bibliometric approximation to the multilevel models, in order to contribute indicators of the general impact that these models have, regarding theoretical as much as applied works. The works that have been reviewed are included in the magazine "Newsletter Multilevel Modelling" during the period of 1989-2005. In the same way, different denominations in which these models appear in the literature have been noted, providing information from diverse areas: educational, psychological and of other sciences. Referring to the use of the telematic networks, certain indispensable article references are indicated to update and follow-up the development and application of these analysis strategies. El presente trabajo es una aproximación bibliométrica a los modelos multinivel, con la idea de aportar indicadores sobre el impacto general que estos modelos tienen, tanto en lo que se refiere a trabajos teóricos como aplicados. En este sentido se revisan los trabajos incluidos en la revista “Newsletter Multilevel Modelling” durante el periodo de 1989-2005. Igualmente, se apuntan las distintas denominaciones en las que los modelos multivariados se presentan en la literatura, proporcionando datos de diversos ámbitos: educativo, psicológico y de otras ciencias. En relación al uso de las redes telemáticas, se señalan algunas referencias imprescindibles para la puesta al día y el seguimiento del desarrollo y aplicación de estas estrategias de análisis.

The association between income inequality and dental caries on early childhood in Colombia was evaluated using a multi-level analysis. We analyzed data from the latest national oral survey (2014) and information about income in absolute and relative terms on a state-level. The outcomes were caries experience, and untreated caries. A multilevel logistic regression model was used (2 levels) with children/households nested within states. Age, gender, area-level socioeconomic position (SEP), household income and health insurance regime were the level 1 explanatory variables. For level 2, variables were the Gini coefficient, Unsatisfied Basic Needs (UBN) and Gross Domestic Product (GDP). Data from 5.250 children, aged 1, 3 and 5 years were evaluated. Prevalence of caries experience and untreated caries was 36.9% and 33.0% respectively. Both outcomes showed significant associations with age, low SEP and belonging to the subsidized health insurance regime: untreated dental caries was associated with living in low and very low SEP (OR: 1.72; 95%CI 1.42, 2.07 and OR: 1.69; 95%CI 1.36, 2.09 respectively), and subsidized health insurance scheme (OR: 1.58; 95%CI 1.11, 2.24). When the Gini, GDP and UBN indicators were included in the models, no significant associations were found.

Este trabajo relaciona el rendimiento obtenido por los alumnos españoles en la evaluación PISA 2006 con variables del entorno del estudiante y del centro que están asociadas a la función básica de producción educativa (Hanushek, 1989) y, por lo tanto, a la eficiencia del sistema educativo. El análisis se ha llevado a cabo mediante la técnica de análisis multinivel, ya que permite el estudio adecuado de estructuras de datos anidados, como ocurre en el caso de datos educativos. A la luz de los resultados obtenidos es posible evidenciar una relación entre las diferentes variables de recursos, tanto de alumno (por ejemplo recursos educativos, posesiones culturales en casa) como de centro (titularidad o número de ordenadores por alumno, entre otras) con el rendimiento académico. This paper connects Spanish students performance in science in PISA 2006 with indicators of students, teachers and institutions associated to the educational production function (Hanushek, 1989) and, consequently, with the efficiency of the educational system. The analysis has been carried out by means of multilevel analysis; this technique is appropriate for the study of hierarchical data structures, as is the case for educational information. In light of the information provided it is possible to demonstrate a relationship between resources variables, both in the student's level (for example educational resources, cultural possessions in home) and at the school level (for example ownership, computers) with the academic performance.