Calibration and XGBoost reweighting to reduce coverage and non-response biases in overlapping panel surveys: application to the Healthcare and Social Survey.

BACKGROUND: Surveys have been used worldwide to provide information on the COVID-19 pandemic impact so as to prepare and deliver an effective Public Health response. Overlapping panel surveys allow longitudinal estimates and more accurate cross-sectional estimates to be obtained thanks to the larger sample size. However, the problem of non-response is particularly aggravated in the case of panel surveys due to population fatigue with repeated surveys. OBJECTIVE: To develop a new reweighting method for overlapping panel surveys affected by non-response. METHODS: We chose the Healthcare and Social Survey which has an overlapping panel survey design with measurements throughout 2020 and 2021, and random samplings stratified by province and degree of urbanization. Each measurement comprises two samples: a longitudinal sample taken from previous measurements and a new sample taken at each measurement. RESULTS: Our reweighting methodological approach is the result of a two-step process: the original sampling design weights are corrected by modelling non-response with respect to the longitudinal sample obtained in a previous measurement using machine learning techniques, followed by calibration using the auxiliary information available at the population level. It is applied to the estimation of totals, proportions, ratios, and differences between measurements, and to gender gaps in the variable of self-perceived general health. CONCLUSION: The proposed method produces suitable estimators for both cross-sectional and longitudinal samples. For addressing future health crises such as COVID-19, it is therefore necessary to reduce potential coverage and non-response biases in surveys by means of utilizing reweighting techniques as proposed in this study.

Impact of COVID-19 on the Health of the General and More Vulnerable Population and Its Determinants: Health Care and Social Survey-ESSOC, Study Protocol.

This manuscript describes the rationale and protocol of a real-world data (RWD) study entitled Health Care and Social Survey (ESSOC, Encuesta Sanitaria y Social). The study's objective is to determine the magnitude, characteristics, and evolution of the COVID-19 impact on overall health as well as the socioeconomic, psychosocial, behavioural, occupational, environmental, and clinical determinants of both the general and more vulnerable population. The study integrates observational data collected through a survey using a probabilistic, overlapping panel design, and data from clinical, epidemiological, demographic, and environmental registries. The data will be analysed using advanced statistical, sampling, and machine learning techniques. The study is based on several measurements obtained from three random samples of the Andalusian (Spain) population: general population aged 16 years and over, residents in disadvantaged areas, and people over the age of 55. Given the current characteristics of this pandemic and its future repercussions, this project will generate relevant information on a regular basis, commencing from the beginning of the State of Alarm. It will also establish institutional alliances of great social value, explore and apply powerful and novel methodologies, and produce large, integrated, high-quality and open-access databases. The information described here will be vital for health systems in order to design tailor-made interventions aimed at improving the health care, health, and quality of life of the populations most affected by the COVID-19 pandemic.

Industrial contaminated sites and health: results of a European survey.

BACKGROUND: industrially contaminated sites (ICSs) have been recognised as a major public health concern since they involve exposure to multiple environmental stressors, normally distributed unevenly within population. The COST Action on Industrially Contaminated Sites and Health Network (ICSHNet) comprises a European network of experts and institutions to clarify needs and priorities for better characterising the impact on environment and health of ICS. OBJECTIVES: evaluate the availability of information and studies concerning selected ICSs in participating Countries within the ICSHNet, with particular consideration on the accessibility to environmental, health and demographic data, and research and assessment tools. METHODS: to evaluate the availability of data, an Action Questionnaire (AQ) was developed based on previous questionnaires used in different European projects and on expert consultation. The AQ, with 84 items organised in eight sections, was adapted to an on-line version using the software LimeSurvey. The survey was sent to 47 participants within the ICSHNet, to report over a list of 99 ICSs previously identified. RESULTS: information was gathered from 81 sites out of the initially selected 99, reported by 45 participants from 27 Countries (82% of Countries in the ICSHNet). The predominant polluting activities were waste disposal (46%) and chemical industries (37%), affecting all environmental media, but more extensively surface and groundwater (70%) and soil (68%). Main categories of contaminants affecting different media were heavy metals and chlorinated hydrocarbons, but also BTEX (benzene, toluene, ethylbenzene, and xylene) and ambient air pollutants (e.g., particulate matter, SOx). Human health risk assessment was the most prevalent methodological approach for characterising impacts on health (32%), followed by epidemiological studies (26%), and health impact assessment (12%). The low reporting, both referring to data availability or methodologies, could be due to absence of data, or to the fact that the reporting person (many of them from the public health sector) did not know how to reach the environmental information. CONCLUSIONS: survey findings suggest that improving the collection and access to specific environmental, health and demographic data related to ICSs is crucial to meet the methodological requirement to better analyse the health impact of ICSs.

La salud de las personas adultas afectadas por un proceso de desahucio / The health of adults undergoing an eviction process

Objetivo: Analizar el estado de salud percibida, y otros indicadores relacionados con la salud, en personas adultas de Granada que se encuentran en un proceso de desahucio de su vivienda habitual, de alquiler o en propiedad, en comparación con la salud de la población general andaluza. Métodos: Estudio transversal mediante encuesta administrada por personal entrenado que incluye instrumentos de la Encuesta Andaluza de Salud 2011 para la medición de variables de salud física y mental, y de hábitos relacionados con la salud. Se han comparado los resultados con los obtenidos sobre la población general andaluza mediante la Encuesta Andaluza de Salud. Se ha realizado un análisis bivariado utilizando la prueba de ji al cuadrado, y un análisis multivariado mediante regresión logística. Resultados: Se ha obtenido una muestra total de 205 personas en proceso de desahucio. El 59,5% (122) son mujeres y el 40,5% (83) hombres. Presentan una mayor probabilidad de tener una salud deficiente (odds ratio [OR]: 12,63; intervalo de confianza del 95% [IC95%]: 8,74-18,27), enfermedad cardiovascular (OR: 3,08; IC95%:1,54-6,16) o consumir tabaco (OR: 1,68; IC95%: 1,21-2,33), en comparación con la población general andaluza. La mayoría de los indicadores analizados muestran un peor resultado para las mujeres que experimentan un proceso de desahucio. Conclusiones: Nuestros resultados indican que las personas afectadas por un proceso de desahucio en Granada y su área metropolitana, en el actual contexto de crisis, expresan una peor salud en relación a la población general andaluza. Es necesario seguir investigando sobre la salud y los desahucios, desde diferentes aproximaciones metodológicas, para una mejor comprensión de este problema (AU)

Objective: To analyze perceived health status and other health-related indicators in the adult population in Granada (Spain) undergoing an eviction process from their homes, whether rented or owned, in comparison with health indicators in the general adult population in Andalusia. Methods: A cross-sectional survey was administered by trained staff. The survey included instruments from the Andalusian Health Survey 2011 for measuring variables related to physical and mental health, as well as health-related habits. We compared the results with those obtained from the Andalusian general population through the Andalusian Health Survey. A bivariate analysis using the ji2 test and a multivariate logistic regression analysis were conducted. Results: We obtained a total sample of 205 people in the process of eviction. A total of 59.5% (n=122) were women, and 40.5% (n=83) were men. Participants were more likely to have poor health (odds ratio [OR]: 12.63, 95% confidence interval [95%CI]: 8.74-18.27), have cardiovascular diseases (OR: 3.08; 95%CI: 1.54- 6.16), or to smoke (OR: 1.68; 95% CI: 1.21-2.33) compared with the Andalusian general population. Most of the health indicators analyzed showed a worse outcome for women undergoing an eviction process. Conclusions: Our results suggest that, in the current context of economic crisis, people undergoing a process of eviction in Granada and its metropolitan area show poorer health than the Andalusian general population. Further research is needed on health and evictions from different methodological approaches, for a better understanding of the topic (AU)

[The health of adults undergoing an eviction process]. / La salud de las personas adultas afectadas por un proceso de desahucio.

OBJECTIVE: To analyze perceived health status and other health-related indicators in the adult population in Granada (Spain) undergoing an eviction process from their homes, whether rented or owned, in comparison with health indicators in the general adult population in Andalusia. METHODS: A cross-sectional survey was administered by trained staff. The survey included instruments from the Andalusian Health Survey 2011 for measuring variables related to physical and mental health, as well as health-related habits. We compared the results with those obtained from the Andalusian general population through the Andalusian Health Survey. A bivariate analysis using the χ2 test and a multivariate logistic regression analysis were conducted. RESULTS: We obtained a total sample of 205 people in the process of eviction. A total of 59.5% (n=122) were women, and 40.5% (n=83) were men. Participants were more likely to have poor health (odds ratio [OR]: 12.63, 95% confidence interval [95%CI]: 8.74-18.27), have cardiovascular diseases (OR: 3.08; 95%CI: 1.54- 6.16), or to smoke (OR: 1.68; 95% CI: 1.21-2.33) compared with the Andalusian general population. Most of the health indicators analyzed showed a worse outcome for women undergoing an eviction process. CONCLUSIONS: Our results suggest that, in the current context of economic crisis, people undergoing a process of eviction in Granada and its metropolitan area show poorer health than the Andalusian general population. Further research is needed on health and evictions from different methodological approaches, for a better understanding of the topic.

Efectos de la edad, el periodo de defunción y la cohorte de nacimiento en la mortalidad por enfermedad isquémica del corazón en el sur de España / Age, period, and cohort effects on mortality from ischemic heart disease in southern Spain

Introducción y objetivos: La cardiopatía isquémica es la primera causa de muerte y una de las cuatro principales causas de carga de enfermedad en el mundo. El objetivo de este trabajo es evaluar los efectos edad-periodo-cohorte la mortalidad por cardiopatía isquémica en Andalucía (sur de España) y en cada una de sus provincias durante el periodo 1981-2008. Métodos: Se realizó un estudio ecológico poblacional. Se analizaron las 145.539 defunciones por cardiopatía isquémica ocurridas en Andalucía durante el periodo de estudio a edades comprendidas entre 30 y 84 años. Se estimó un modelo de regresión no lineal con funciones spline para cada sexo y área geográfica. Resultados: En la tasa de mortalidad masculina y femenina a partir de 30 años de edad se observa tendencia a aumentar. El riesgo de muerte para varones y mujeres fue descendiente para las cohortes nacidas después de 1920 y con una pendiente más pronunciada después de 1960 entre los varones. El análisis del efecto periodo mostró que el riesgo de mortalidad masculina y femenina se mantuvo estable desde 1981 hasta 1990, aumentó entre 1990 y 2000 y volvió a disminuir desde 2000 hasta 2008. Conclusiones: Los efectos edad-periodo-cohorte en la mortalidad fueron similares en todas las provincias de Andalucía y el conjunto de la comunidad autónoma. Si los efectos cohorte y periodo persisten, la mortalidad masculina y femenina por enfermedad isquémica cardiaca continuará disminuyendo (AU)

Introduction and objectives: Ischemic heart disease is the leading cause of death and one of the top 4 causes of burden of disease worldwide. The aim of this study was to evaluate age-period-cohort effects on mortality from ischemic heart disease in Andalusia (southern Spain) and in each of its 8 provinces during the period 1981-2008. Methods: A population-based ecological study was conducted. In all, 145 539 deaths from ischemic heart disease were analyzed for individuals aged between 30 and 84 years who died in Andalusia in the study period. A nonlinear regression model was estimated for each sex and geographical area using spline functions. Results: There was an upward trend in male and female mortality rate by age from the age of 30 years. The risk of death for men and women showed a downward trend for cohorts born after 1920, decreasing after 1960 with a steep slope among men. Analysis of the period effect showed that male and female death risk first remained steady from 1981 to 1990 and then increased between 1990 and 2000, only to decrease again until 2008. Conclusions: There were similar age-period-cohort effects on mortality in all the provinces of Andalusia and for Andalusia as a whole. If the observed cohort and period effects persist, male and female mortality from ischemic heart disease will continue to decline (AU)

Age, Period, and Cohort Effects on Mortality From Ischemic Heart Disease in Southern Spain.

INTRODUCTION AND OBJECTIVES: Ischemic heart disease is the leading cause of death and one of the top 4 causes of burden of disease worldwide. The aim of this study was to evaluate age-period-cohort effects on mortality from ischemic heart disease in Andalusia (southern Spain) and in each of its 8 provinces during the period 1981-2008. METHODS: A population-based ecological study was conducted. In all, 145 539 deaths from ischemic heart disease were analyzed for individuals aged between 30 and 84 years who died in Andalusia in the study period. A nonlinear regression model was estimated for each sex and geographical area using spline functions. RESULTS: There was an upward trend in male and female mortality rate by age from the age of 30 years. The risk of death for men and women showed a downward trend for cohorts born after 1920, decreasing after 1960 with a steep slope among men. Analysis of the period effect showed that male and female death risk first remained steady from 1981 to 1990 and then increased between 1990 and 2000, only to decrease again until 2008. CONCLUSIONS: There were similar age-period-cohort effects on mortality in all the provinces of Andalusia and for Andalusia as a whole. If the observed cohort and period effects persist, male and female mortality from ischemic heart disease will continue to decline.

[Referrals and sex of physicians and patients: a gender analysis, Andalussia, Spain]. / Derivaciones en los centros de salud de Andalucía según el sexo de profesionales y pacientes: un análisis de género.

BACKGROUND: It has been identified differences of medical care practice in primary care related to physician's sex. Simultaneously, there are gender inequalities in the assignment of health resources. Both aspects give rise to an increasing growing interest in the management and provision of health services. OBJECTIVES: To examine the differences in the referral practice made by female and male primary care physicians working in health centers in Andalusia, to consider whether there are disparities in referrals received by men and women, and to examine the interaction between patient's sex and physician's sex. METHODS: Observational, cross-sectional, and multicenter study. POPULATION: 4 health districts in Andalucía and their physicians. SAMPLE: 382 physicians. MEASUREMENTS: referral rate per visit (RV), referral rate per patient quota (RQ), patient's sex, physician: sex, age, postgraduate family medicine specialty, size of the patient quota by sex, mean number of patients/day by sex, mean age of the patient quota by sex, and proportion of men in the quota. Health center: urban / rural, size of the team, enrolled population, and postgraduate family medicine specialty's accreditation. SOURCES: databases of health districts. PERIOD OF STUDY: 2010. ANALYSIS: Bivariate and multivariate multilevel analysis of the referral rate per visit with mixed Poisson model. RESULTS: In 2010 382 physicians made 129,161 referrals to specialized care. The RQ was 23.47 and the RV was 4.92. The RQ in women and men was 27.23 and 19.78 for women physicians, being 27.37 and 19.51 for male physicians. The RV in women and men was 4.92 and 5.48 for women physicians, being 4.54 and 4.93 for male physicians. CONCLUSION: There are no differences in referral according to physician's sex. However, there are signs that might indicate the existence of gender inequality, and women patient received less referrals. There are no physician-patient's sex interaction.

Derivaciones en los centros de salud de Andalucía según el sexo de profesionales y pacientes. Un análisis de género / Referrals and sex of physicians and patients. A gender analysis, Andalussia, Spain

Fundamento: En atención primaria se han identificado diferencias de práctica según sexo del profesional y, simultáneamente, existen des- igualdades de género en la asignación de recursos sanitarios, aspectos ambos que suscitan un interés creciente en la gestión y provisión de servicios de salud. El objetivo del estudio es conocer si existen diferencias de práctica en las derivaciones sanitarias realizadas por médicas y médicos de familia (MF) de centros de salud de Andalucía, si existen desigualdades en las derivaciones recibidas por hombres y mujeres, y si existe inter- acción sexo de profesional sexo de paciente. Métodos: Estudio transversal y multicéntrico. Población: MF de 4 distritos sanitarios (DDSS). Muestra: 382 MF. Variables: tasa de derivaciones por visita (TDxV), tasa de derivaciones por cupo (TDxC), sexo de paciente; sexo, edad, y formación postgraduada en medicina familiar de MF, tamaño del cupo por sexo, media de visitas /paciente por sexo, edad media del cupo por sexo, y proporción de hombres en el cupo; centro de salud urbano/rural, tamaño del equipo, población adscrita y acreditación docente. Fuentes: bases de datos de los DDSS. Análisis estadístico: descriptivo. Bivariante y multivariante mediante análisis multinivel de la TDxV con modelo mixto de Poisson. Resultados: En 2010 los/as 382 MF realizaron 129.161 derivaciones a especialistas. La TDxC fue 23,47 y la TDxV es 4,92. Las TDxC de las médicas fue 27,23 en mujeres y 19,78 en hombres y las de los médicos 27,37 en mujeres y 19,51 en hombres. La TDxV de las médicas fueron 4,92 en mujeres y 5,48 en hombres y para los médicos 4,54 y 4,93 respectivamente. Conclusiones: No existen diferencias en las derivaciones según sexo de las mujeres son menos derivadas. No existe interacción sexo profesional-sexo paciente (AU)

Background: It has been identified differences of medical care practice in primary care related to physician’s sex. Simultaneously, there are gender inequalities in the assignment of health resources. Both aspects give rise to an increasing growing interest in the management and provi- sion of health services. Objectives: To examine the differences in the referral practice made by female and male primary care physicians working in health centers in Andalusia, to consider whether there are disparities in referrals received by men and women, and to examine the interaction bet- ween patient’s sex and physician’s sex. Methods: Observational, cross-sectional, and multicenter study. Population: 4 health districts in Andalucía and their physicians. Sample: 382 physicians. Measurements: referral rate per visit (RV), referral rate per patient quota (RQ), patient's sex, physician: sex, age, postgraduate family medicine specialty, size of the patient quota by sex, mean number of patients/day by sex, mean age of the patient quota by sex, and proportion of men in the quota. Health center: urban / rural, size of the team, enrolled population, and postgraduate family medicine specialty's accreditation. Sources: databases of health districts. Period of study: 2010. Analysis: Bivariate and multivariate multilevel analysis of the referral rate per visit with mixed Poisson model. Results: In 2010 382 physicians made 129,161 referrals to specialized care. The RQ was 23.47 and the RV was 4.92. The RQ in women and men was 27.23 and 19.78 for women physicians, being 27.37 and 19.51 for male physicians. The RV in women and men was 4.92 and 5.48 for women physicians, being 4.54 and 4.93 for male physicians. Conclusion: There are no differences in referral according to physician´s sex. However, there are signs that might indicate the existence of men patient received less referrals. There are no physician-patient's sex interaction (AU)

Incorporación de la prescripción electrónica en un distrito de atención primaria: implicaciones en el gasto farmacéutico y factores determinantes de su utilización / Introduction of the electronic prescription in a primary care district: Impact on pharmaceutical expenditure and the factors determining its use

OBJETIVO: Analizar las implicaciones del uso de receta electrónica sobre el gasto farmacéutico (GF) así como explorar factores determinantes de su utilización. MÉTODOS: Construcción de 2 modelos de regresión múltiple, tomando en uno de ellos como variable dependiente el GF y en el otro el porcentaje de prescripción mediante receta electrónica, incluyendo datos referidos a 9 centros de atención primaria de Andalucía. RESULTADOS: Cada incremento en un punto de prescripción por principio activo (PPA) se relacionó con una reducción media del GF de 2.079 €/año, mientras que cada incremento porcentual en el uso de receta electrónica se relacionó con un aumento de 1.027 € anuales. También se relacionaron con un mayor GF cada tarjeta ajustada por edad y tipo de farmacia (TAFE) asignada (36,71 €/año), la presión asistencial (2.585 €/año) y la frecuentación (2.17.148 €/año). Los factores que se relacionaron con un mayor uso de receta electrónica fueron un menor índice de ruralidad, una mayor distancia a la capital, una mayor PPA y una menor frecuentación. CONCLUSIONES: La prescripción mediante receta electrónica parece ser una variable predictora del gasto farmacéutico, relacionándose cada incremento porcentual con un incremento medio anual de 1.027 euros. Esta variable, junto con la prescripción mediante principio activo, el número de TAFE asignadas, la frecuentación y la presión asistencial media, contribuyen a explicar más del 65% de la variabilidad en el GF entre facultativos. El incremento del gasto relacionado con la receta electrónica parece producirse a expensas del aumento en el número de prescripciones por paciente

OBJECTIVE: To analyse the impact of using the electronic prescription on pharmaceutical expenditure (PE), and to explore the factors determining its use. METHODS: Two multiple regression models were constructed. For these two models, the PE and the percentage of electronic prescription use were taken respectively as dependent variables. Data from nine Primary care centres were included. RESULTS: Each point increase in prescription by active principle (PAP) was associated with an average reduction of 2,079 PE € per year. In contrast, every percentage increase in the use of electronic prescriptions was associated with an annual increase of 1,027 €. In addition, factors like card adjusted for age and type of pharmacy (TAFE) allocated (€ 36.71/year), the burden of care (€ 2,585/year) and frequency of clinic visits (€ 17,148/year), were also associated with increased PE. The other factors associated with an increase in the electronic prescriptions were, lower rurality rates, greater distance from the capital, greater PAP, as well as less clinic visits. CONCLUSIONS: Electronic prescription use seems to be a predictor of pharmaceutical expenditure in which every percentage increase relates to an average annual increase of 1,027 Euros. This variable along with the prescription by active principle, and the amount of TAFEs assigned, as well as the burden of care and number of visits, are variables which help to explain more than 65% of the PE variability between physicians. Likewise, the increase in expenditures related to electronic prescriptions seems to occur at the expense of increasing number of prescriptions per patient

[Introduction of the electronic prescription in a primary care district: impact on pharmaceutical expenditure and the factors determining its use]. / Incorporación de la prescripción electrónica en un distrito de atención primaria: implicaciones en el gasto farmacéutico y factores determinantes de su utilización.

OBJECTIVE: To analyse the impact of using the electronic prescription on pharmaceutical expenditure (PE), and to explore the factors determining its use. METHODS: Two multiple regression models were constructed. For these two models, the PE and the percentage of electronic prescription use were taken respectively as dependent variables. Data from nine Primary care centres were included. RESULTS: Each point increase in prescription by active principle (PAP) was associated with an average reduction of 2,079 PE € per year. In contrast, every percentage increase in the use of electronic prescriptions was associated with an annual increase of 1,027 €. In addition, factors like card adjusted for age and type of pharmacy (TAFE) allocated (€ 36.71/year), the burden of care (€ 2,585/year) and frequency of clinic visits (€ 17,148/year), were also associated with increased PE. The other factors associated with an increase in the electronic prescriptions were, lower rurality rates, greater distance from the capital, greater PAP, as well as less clinic visits. CONCLUSIONS: Electronic prescription use seems to be a predictor of pharmaceutical expenditure in which every percentage increase relates to an average annual increase of 1,027 Euros. This variable along with the prescription by active principle, and the amount of TAFEs assigned, as well as the burden of care and number of visits, are variables which help to explain more than 65% of the PE variability between physicians. Likewise, the increase in expenditures related to electronic prescriptions seems to occur at the expense of increasing number of prescriptions per patient.

Uncontrolled migrations as a cause of inequality in health and mortality in small-area studies.

Mortality is one of the most widely used indicators in small-area ecologic studies. Both accessibility to mortality data and advances in the development of new disease mapping techniques have contributed to an abundance of mortality maps and atlases over the last decade. Results may be biased in this kind of study if there has been unmeasured geographic mobility of the population. Most published papers tend to neglect this possibility. We use the theory of dynamics systems to demonstrate that migratory flows unmonitored by official population registers may lead to major errors in mortality rates and relative risks. Simulations in 4 scenarios showed more than 8% underestimation of the mortality rate and more than 11% underestimation of relative risk in areas with high uncontrolled emigration, and above 19% overestimation of mortality rate and above 15% overestimation of relative risk in areas with high uncontrolled immigration.Researchers conducting small-area epidemiologic studies should explore the reliability of population information in geographic areas before drawing hypothesis or conclusions on other possible causes of mortality differences.

[Interactive mortality atlas in Andalusia, Spain (AIMA)]. / Atlas interactivo de mortalidad en Andalucía (AIMA).

Until now, mortality atlases have been static. Most of them describe the geographical distribution of mortality using count data aggregated over time and standardized mortality rates. However, this methodology has several limitations. Count data aggregated over time produce a bias in the estimation of death rates. Moreover, this practice difficult the study of temporal changes in geographical distribution of mortality. On the other hand, using standardized mortality hamper to check differences in mortality among groups. The Interactive Mortality Atlas in Andalusia (AIMA) is an alternative to conventional static atlases. It is a dynamic Geographical Information System that allows visualizing in web-site more than 12.000 maps and 338.00 graphics related to the spatio-temporal distribution of the main death causes in Andalusia by age and sex groups from 1981. The objective of this paper is to describe the methods used for AIMA development, to show technical specifications and to present their interactivity. The system is available from the link products in www.demap.es. AIMA is the first interactive GIS that have been developed in Spain with these characteristics. Spatio-temporal Hierarchical Bayesian Models were used for statistical data analysis. The results were integrated into web-site using a PHP environment and a dynamic cartography in Flash. Thematic maps in AIMA demonstrate that the geographical distribution of mortality is dynamic, with differences among year, age and sex groups. The information nowadays provided by AIMA and the future updating will contribute to reflect on the past, the present and the future of population health in Andalusia.

Atlas interactivo de mortalidad en Andalucía (AIMA) / Interactive mortality atlas in Andalusia, Spain (AIMA)

Los atlas de mortalidad publicados hasta la fecha son estáticos. La mayoría describen la distribución geográfica de la mortalidad agrupando datos de varios años y usando razones de mortalidad estandarizadas que resumen la mortalidad de todos los grupos de edad. Sin embargo, esta metodología presenta limitaciones para estudiar cambios temporales en la distribución geográfica de la mortalidad o diferencias entre grupos etarios. Aplicando una metodología alternativa surge el Atlas Interactivo de Mortalidad en Andalucía (AIMA), un Sistema de Información Geográfica dinámico que permite visualizar en web-site más de 12.000 mapas y 338.000 gráficos de tendencias correspondientes a la distribución espacio-temporal de las principales causas de muerte en Andalucía por grupos de edad y sexo desde 1981. El objetivo de este trabajo es describir la metodología utilizada para el desarrollo de AIMA, dar a conocer sus características y especificaciones técnicas y mostrar las posibilidades interactivas que ofrece. El sistema está disponible en el hipervínculo productos de www.demap.es. AIMA es el primer SIG interactivo de estas características implementado en España. Para el análisis de datos se han utilizado modelos jerárquicos bayesianos espacio-temporales, integrando posteriormente los resultados en web-site utilizando entorno PHP y cartografía dinámica en formato Flash. La selección de mapas temáticos muestra el carácter dinámico de la distribución geográfica de la mortalidad, con un patrón diferente para cada año, grupo de edad y sexo. La información contenida actualmente en este sistema y las futuras actualizaciones contribuirán a la reflexión sobre el pasado, presente y futuro de la salud en Andalucía (AU)

Until now, mortality atlases have been static. Most of them describe the geographical distribution of mortality using count data aggregated over time and standardized mortality rates. However, this methodology has several limitations. Count data aggregated over time produce a bias in the estimation of death rates. Moreover, this practice difficult the study of temporal changes in geographical distribution of mortality. On the other hand, using standardized mortality hamper to check differences in mortality among groups. The Interactive Mortality Atlas in Andalusia (AIMA) is an alternative to conventional static atlases. It is a dynamic Geographical Information System that allows visualizing in web-site more than 12.000 maps and 338.00 graphics related to the spatio-temporal distribution of the main death causes in Andalusia by age and sex groups from 1981. The objective of this paper is to describe the methods used for AIMA development, to show technical specifications and to present their interactivity. The system is available from the link products in www.demap.es. AIMA is the first interactive GIS that have been developed in Spain with these characteristics. Spatio-temporal Hierarchical Bayesian Models were used for statistical data analysis. The results were integrated into web-site using a PHP environment and a dynamic cartography in Flash. Thematic maps in AIMA demonstrate that the geographical distribution of mortality is dynamic, with differences among year, age and sex groups. The information nowadays provided by AIMA and the future updating will contribute to reflect on the past, the present and the future of population health in Andalusia (AU)

Rural habitat and risk of death in small areas of Southern Spain.

The purpose of this paper is to study the linkage between type of habitat and mortality from all causes in small areas of Southern Spain. An ecological study was conducted on 99,870 people who died between 1985 and 1999. The municipality was taken as the unit of analysis. Data analysis was carried out through hierarchical spatio-temporal bayesian models. Results show a 13.3% reduction in mortality rates among men and 14.1% among women in the most rural areas compared to more urban environments. The study demonstrates the usefulness of socio-demographic indices in small-area geographical analyses.

Protocolo de investigación del atlas de mortalidad de las capitales de provincia de Andalucía y Cataluña (Proyecto AMCAC) / Research protocol for the mortality atlas of the provincial capitals of Andalusia and Catalonia (AMCAC Project)

El propósito de este trabajo es dar a conocer el proyecto multicéntricoAMCAC, que tiene como objetivos describir la distribucióngeográfica de la mortalidad por todas las causas en las secciones censalesde las capitales de provincia de Andalucía y Cataluña durante elperiodo 1992-2002 y 1994-2000 respectivamente, así como estudiarla relación entre las características sociodemográficas de las seccionescensales y la mortalidad.Es un estudio ecológico donde la unidad de análisis es la seccióncensal. Se analizarán los datos relativos a 298.731 individuos(152.913 hombres y 145.818 mujeres) fallecidos en las ciudades deAlmería, Barcelona, Cádiz, Córdoba, Girona, Granada, Huelva,Jaén, Lleida, Málaga, Sevilla y Tarragona durante los periodos deestudio.La variable dependiente es el número de muertes observadas porsección censal. Las variables independientes son el porcentaje dedesempleo, de analfabetismo y de trabajadores manuales. La estimacióndel riesgo relativo suavizado y el estudio de la relación entre lascaracterísticas sociodemográficas de las secciones censales y la mortalidadse realizará para cada ciudad y sexo mediante el modeloBesag-York-Mollié.La difusión de los resultados ayudará a mejorar y ampliar losconocimientos sobre la salud de la población, siendo un punto departida importante para conocer la influencia de variables contextualesen la salud de la población urbana

The aim of this work is to make known the multicentric projectAMCAC, whose objective is to describe the geographical distributionof mortality from all causes in census groups of the provincialcapitals of Andalusia and Catalonia during 1992-2002 and 1994-2000 respectively, and to study the relationship between the sociodemographiccharacteristics of the census groups and mortality. This isan ecological study in which the analytical unit is the census group.The data correspond to 298,731 individuals (152,913 men and145,818 women) who died during the study periods in the towns ofAlmeria, Barcelona, Cadiz, Cordoba, Girona, Granada, Huelva,Jaen, Lleida, Malaga, Seville and Tarragona during the studyperiods. The dependent variable is the number of deaths observedper census group. The independent variables are the percentage ofunemployment, illiteracy and manual workers. Estimation of themoderated relative risk and the study of the associations among thesociodemographic characteristics of the census groups and the mortalitywill be done for each town and each sex using the Besag-York-Mollie model. Dissemination of the results will help to improve andbroaden knowledge about the population's health, and will providean important starting point to establish the influence of contextualvariables on the health of urban populations

Socio-economic level, farming activities and risk of cancer in small areas of Southern Spain.

An ecological design was used to study the relationship between cancer incidence and both socioeconomic and environmental features in Southern Spain. Twenty-four sites and 26,380 cases diagnosed in 1985--1996 were analysed. Generalised Additive Models were used for data analysis. Except for lip cancer, the urban areas showed an increase in cancer risk for all sites. The relative risks among urban and rural municipalities ranges between 1.09 for skin non-melanoma (95% CI: 1.00-1.18) and 1.64 for cervix cancer (95% CI: 1.28-2.12). The relative risk among areas with high and low unemployment was 1.29 for stomach cancer (95% CI: 1.07-1.57), 1.45 for oral cavity cancer (95% CI: 1.10-1.93) and 1.77 for oesophagus cancer (95% CI: 1.02-3.05). Areas with highest unemployment showed the lowest incidence of melanoma. Risk for leukaemia, gall bladder, breast and prostate cancer showed a significant decreases by approximately 28% in the municipalities with the highest illiteracy score. A high percentage of land under cultivation was related to uterine tumours, larynx, rectum, lung, skin non-melanoma and brain cancers. For these sites, the risk had a significant increase by between 23% (skin non-melanoma) and 70% (rectum). Areas with high intensive farming showed a significant increase in cancer risk for lip, oral cavity, larynx, oesophagus, colon, lung, and bladder cancer. The relative risks ranges between 1.16 for colon cancer (95% CI: 1.04-1.29) and 1.47 for oesophagus cancer (95% CI: 1.15-1.87). The results of this study reveal how important socio-economic and environmental factors are for the analysis of cancer incidence in small areas of Southern Spain.