Emanuela Dreassi | Università degli Studi di Firenze (University of Florence) (original) (raw)
Papers by Emanuela Dreassi
This paper posits that the municipality level offers important insights into the study of tempora... more This paper posits that the municipality level offers important insights into the study of temporal and spatial patterns of family change. We focus on the diffusion of one-parent families in Italy: variation in the structure of co-resident domestic groups is a crucial indicator of changing diversity in family patterns. We apply a hierarchical Bayesian spatio-temporal model to the data of the last three Italian Population Censuses, at the municipality level. Our results show substantial sub-regional and sub-provincial heterogeneities in the spatial organization of family systems. These patterns might have gone undetected if larger territorial units of analysis had been considered.
Bernoulli, 2021
In a Bayesian framework, to make predictions on a sequence X1, X2,. .. of random observations, th... more In a Bayesian framework, to make predictions on a sequence X1, X2,. .. of random observations, the inferrer needs to assign the predictive distributions σn(•) = P Xn+1 ∈ • | X1,. .. , Xn. In this paper, we propose to assign σn directly, without passing through the usual prior/posterior scheme. One main advantage is that no prior probability has to be assessed. The data sequence (Xn) is assumed to be conditionally identically distributed (c.i.d.) in the sense of [4]. To realize this programme, a class Σ of predictive distributions is introduced and investigated. Such a Σ is rich enough to model various real situations and (Xn) is actually c.i.d. if σn belongs to Σ. Furthermore, when a new observation Xn+1 becomes available, σn+1 can be obtained by a simple recursive update of σn. If µ is the a.s. weak limit of σn, conditions for µ to be a.s. discrete are provided as well.
Statistics & Probability Letters, 2021
The probability distribution of a sequence X = (X 1 , X 2 ,. . .) of random variables is determin... more The probability distribution of a sequence X = (X 1 , X 2 ,. . .) of random variables is determined by its predictive distributions P(X 1 ∈ •) and P (X n+1 ∈ • | X 1 ,. .. , X n) , n ≥ 1. Motivated by applications in Bayesian predictive inference, in Berti et al. (2020), a class C of sequences is introduced by specifying such predictive distributions. Each X ∈ C is conditionally identically distributed. The asymptotics of X ∈ C is investigated in this paper. Both strong and weak limit theorems are provided. Conditions for X to converge a.s., and for X not to converge in probability, are given in terms of the predictive distributions. A stable CLT is provided as well. Such a CLT is used to obtain approximate credible intervals.
Cancer Management and Research, 2020
Purpose: Lung cancer is considered as a common cause of cancer mortality. The disease represents ... more Purpose: Lung cancer is considered as a common cause of cancer mortality. The disease represents the second and third causes of deaths from cancer among Iranian women and men, respectively. The present study aimed to evaluate the spatial variations in relative risk of lung cancer mortality in Iran and its relation to common risk factors between men and women and specific risk factors among women. Methods: In this ecological study, the lung cancer mortality data were analyzed in Iran during 2011-2014. Besag, York, and Mollie's (BYM) model and shared component model (SCM) were used to compare the spatial variations of the relative risk of lung cancer mortality by applying OpenBUGS version 3.2.3 and R version 3.6.1. Results: The median age for death due to lung cancer in Iran is 74 years. During 2011-2014, the age-standardized lung cancer mortality rates among men and women were 12 and 5 per 100,000 individuals, respectively. In addition, almost similar spatial patterns were observed for both men and women. Further, risk factors, which are shared between men and women, were considered as the main cause of variation of lung cancer mortality relative risk in the regions under study for both men and women. The highest impact of the women-specific risk factors was estimated in northeastern and southwestern of the country while the lowest was related to Gilan province in northern part of Iran. Conclusion: Based on the spatial pattern, lung cancer risk factors are at relatively high levels in most parts of Iran, especially in the northwest of the country. Regarding the women, the high-risk regions were considerably extended. Further, the highest concentration of the specific risk factors among women was observed in the eastern, central, and southwestern parts. The smoking effect, and the second-smoking effect and environmental pollutions could play more significant roles for men and women, respectively.
Aesthetic plastic surgery, Jan 3, 2017
Pectus excavatum (PE) is one of the most frequent thoracic malformations. Generally, the malforma... more Pectus excavatum (PE) is one of the most frequent thoracic malformations. Generally, the malformation is not associated with functional disorders and often constitutes an aesthetic alteration with significant psychological distress. To reduce the visibility of the residual scarring produced by corrective surgery and to improve the aesthetic outcome, the authors propose a new prosthetic implant technique through a periareolar access. From January 2005 to January 2015, 11 patients affected with PE underwent the surgical procedure with a sternal prosthesis implanted through a periareolar access with the help of a fiberscope. The preoperative evaluation of the perception of the malformation and postoperation results were made using different questionnaires. The data collected in our series were compared with that reported in 4 different papers where other forms of access were used: sternal, inframammary and transumbilical. No major complications or dislocation of the implants were repor...
In vivo (Athens, Greece), Jan 2, 2017
Evaluation of long-term results after aponeurotic blepharoptosis correction with external levator... more Evaluation of long-term results after aponeurotic blepharoptosis correction with external levator muscle complex advancement. We carried out a retrospective study with medical record review of 20 patients (40 eyes) affected by bilateral aponeurotic moderate and severe ptosis who underwent primary surgery between January 2010 and December 2013. Criteria for outcome evaluations included 3-year postoperative follow-up of upper margin reflex index (uMRD) and symmetry. 3-Year postoperative follow-up showed 17 (85%) cases of successful correction of ptosis and three cases (15%) showed partial success. Two eyes showed hypocorrection, while one eye was overcorrected. The symmetry was maintained in all patients except for the oldest. External superior levator advancement is an effective procedure for moderate and severe aponeurotic blepharoptosis correction, and establishes good long-term eyelid position and symmetry.
Journal of Plastic, Reconstructive & Aesthetic Surgery, 2017
La pubblicazione è resa disponibile sotto le norme e i termini della licenza di deposito, secondo... more La pubblicazione è resa disponibile sotto le norme e i termini della licenza di deposito, secondo quanto stabilito dalla Policy per l'accesso aperto dell'Università degli Studi di Firenze (https://www.sba.unifi.it/upload/policy-oa-2016-1.pdf)
Viene analizzato il legame tra condizione socio-economica e mortalità per tumore al polmone nella... more Viene analizzato il legame tra condizione socio-economica e mortalità per tumore al polmone nella popolazione maschile della Toscana, a livello comunale, nel periodo 1971-94, facendo uso di un modello spazio-temporale Bayesiano con covariate tempo dipendenti. In particolare si intende indagare se tra la manifestazione di un certo livello di deprivazione materiale e quella della mortalità vi sia o meno un tempo di latenza, quindi se la mortalitàè legata alla deprivazione materiale di dieci anni prima (esposizione a fattori di rischio), oppure a quella presente (problemi di accesso al sistema sanitario).
D ip a r tim e nto d i S ta tis tic a " G. Pa r e nti"-V ia le Mo r g a g ni 5 9-5 0 1 3 4 F ir e... more D ip a r tim e nto d i S ta tis tic a " G. Pa r e nti"-V ia le Mo r g a g ni 5 9-5 0 1 3 4 F ir e nz e-w w w .d s .unif i.it
Riassunto: Analizzare congiuntamente la distribuzione spazio-temporale di una malattia in due gru... more Riassunto: Analizzare congiuntamente la distribuzione spazio-temporale di una malattia in due gruppi disgiunti di popolazione permette di fare inferenza su eventuali andamenti comuni ed effettuare ipotesi sulla condivisione di fattori eziologici. In questo lavoro si analizza la mortalità per tumore al polmone nei Comuni della Toscana per gli uomini e le donne appartenenti alle coorti di nati dal 1905 al 1940. Si utilizza una serie di modelli Bayesiani gerarchici spaziotemporali con componenti di varianza comuni e specifiche per genere.
A new approach to ecological regression for disease mapping is introduced, based on semiparametri... more A new approach to ecological regression for disease mapping is introduced, based on semiparametric M-quantile regression models. In particular, we define a Negative Binomial M-quantile model as an alternative to Empirical Bayes or fully Bayesian approaches to disease mapping. The area-level covariates used in ecological regression are usually measured with error, and the proposed M-quantile modelling approach is easily made robust against outlying data in the model covariates. Differences between the M-quantile model and the usual random effects models are discussed, and these alternative approaches are compared using the well-known Scottish Lip cancer data and a simulation experiment. The Lip Cancer data example shows that the Negative Binomial M-quantile model confirms results obtained by other methods, but also seems to have less shrinkage than the Empirical Bayes method, so reducing the problem of oversmoothing. The simulation experiment suggests that the new model leads to estimates with smaller mean square error. We also show how the Negative Binomial M-quantile can be extended to account for spatial correlation between areas using a Geographically Weighted Regression strategy.
Studies in Classification, Data Analysis, and Knowledge Organization, 2012
Two models for jointly analysing the spatial variation of incidences of three (or more) diseases,... more Two models for jointly analysing the spatial variation of incidences of three (or more) diseases, with common and uncommon risk factors, are compared via a simulation experiment. In both models, the linear predictor can be decomposed into shared and disease-specific spatial variability components (named shared clustering and specific clustering respectively). The two models are the shared model on the original formulation that use exchangeable Poisson distribution as response multivariate variable and shared components model that use a Multinomial one. The simulation study shows that models behave similarly. However, Multinomial shared components model performs better for disease-specific spatial variability clustering terms but it is lower for the shared one.
Encyclopedia of Quality of Life and Well-Being Research, 2014
Statistical Modelling, 2001
Statistical Methods in Medical Research, 2014
A new semiparametric approach to model-based small area prediction for counts is proposed and use... more A new semiparametric approach to model-based small area prediction for counts is proposed and used for estimating the average number of visits to physicians for Health Districts in Central Italy. The proposed small area predictor can be viewed as an outlier robust alternative to the more commonly used empirical plug-in predictor that is based on a Poisson generalized linear mixed model with Gaussian random effects. Results from the real data application and from a simulation experiment confirm that the proposed small area predictor has good robustness properties and in some cases can be more efficient than alternative small area approaches.
M-Quantile spatially structured model. The proposed approach is easily made robust against outlyi... more M-Quantile spatially structured model. The proposed approach is easily made robust against outlying data values for covariates. Robust ecological disease mapping is desirable since covariates at area level usually present measure-type error. We
Environmetrics, 2003
ABSTRACT
Environmetrics, 2003
Joint modelling of space and time variation of the risk of disease is an important topic in descr... more Joint modelling of space and time variation of the risk of disease is an important topic in descriptive epidemiology. Most of the proposals in this field deal with at most two time scales (age–period or age–cohort). We propose a hierarchical Bayesian model that can be used as a general framework to jointly study the evolution in time and the spatial pattern of the risk of disease. The rates are modelled as a function of purely spatial terms (local effects of risk factors that do not vary in time), time effects (on the three time axes: age, calendar period and birth cohort) and space–time interactions that describe area specific time patterns. Copyright © 2003 John Wiley & Sons, Ltd.
Environmental and Ecological Statistics, 2005
In this work we propose a Bayesian ecological analysis in which a latent variable summarizes data... more In this work we propose a Bayesian ecological analysis in which a latent variable summarizes data on emissions of atmospheric pollutants. We specified a hierarchical Bayesian model with spatially structured and unstructured random terms with a nested latent factor model. This can be considered a combination of the convolution spatial model of Besag et al. (1991) and an ecological regression analysis in which a latent variable plays the role of the covariate. The unified approach allows to proper account for the uncertainty in the latent score estimation in the regression analysis. The Bayesian Latent Factor model is used to summarize the information on environmental pressure derived from three stressors: Carbon Monoxide, Nitrogen Oxides and Inhalable Particles. We found evidence of positive correlation between Lung cancer mortality and environmental pressure indicators, in males, Tuscany (Italy), 1995-1999. Environmental pressure seems to be restricted to fourteen municipalities (top 5% of the Latent Factor distribution). The model identified two areas with high point source emissions.
This paper posits that the municipality level offers important insights into the study of tempora... more This paper posits that the municipality level offers important insights into the study of temporal and spatial patterns of family change. We focus on the diffusion of one-parent families in Italy: variation in the structure of co-resident domestic groups is a crucial indicator of changing diversity in family patterns. We apply a hierarchical Bayesian spatio-temporal model to the data of the last three Italian Population Censuses, at the municipality level. Our results show substantial sub-regional and sub-provincial heterogeneities in the spatial organization of family systems. These patterns might have gone undetected if larger territorial units of analysis had been considered.
Bernoulli, 2021
In a Bayesian framework, to make predictions on a sequence X1, X2,. .. of random observations, th... more In a Bayesian framework, to make predictions on a sequence X1, X2,. .. of random observations, the inferrer needs to assign the predictive distributions σn(•) = P Xn+1 ∈ • | X1,. .. , Xn. In this paper, we propose to assign σn directly, without passing through the usual prior/posterior scheme. One main advantage is that no prior probability has to be assessed. The data sequence (Xn) is assumed to be conditionally identically distributed (c.i.d.) in the sense of [4]. To realize this programme, a class Σ of predictive distributions is introduced and investigated. Such a Σ is rich enough to model various real situations and (Xn) is actually c.i.d. if σn belongs to Σ. Furthermore, when a new observation Xn+1 becomes available, σn+1 can be obtained by a simple recursive update of σn. If µ is the a.s. weak limit of σn, conditions for µ to be a.s. discrete are provided as well.
Statistics & Probability Letters, 2021
The probability distribution of a sequence X = (X 1 , X 2 ,. . .) of random variables is determin... more The probability distribution of a sequence X = (X 1 , X 2 ,. . .) of random variables is determined by its predictive distributions P(X 1 ∈ •) and P (X n+1 ∈ • | X 1 ,. .. , X n) , n ≥ 1. Motivated by applications in Bayesian predictive inference, in Berti et al. (2020), a class C of sequences is introduced by specifying such predictive distributions. Each X ∈ C is conditionally identically distributed. The asymptotics of X ∈ C is investigated in this paper. Both strong and weak limit theorems are provided. Conditions for X to converge a.s., and for X not to converge in probability, are given in terms of the predictive distributions. A stable CLT is provided as well. Such a CLT is used to obtain approximate credible intervals.
Cancer Management and Research, 2020
Purpose: Lung cancer is considered as a common cause of cancer mortality. The disease represents ... more Purpose: Lung cancer is considered as a common cause of cancer mortality. The disease represents the second and third causes of deaths from cancer among Iranian women and men, respectively. The present study aimed to evaluate the spatial variations in relative risk of lung cancer mortality in Iran and its relation to common risk factors between men and women and specific risk factors among women. Methods: In this ecological study, the lung cancer mortality data were analyzed in Iran during 2011-2014. Besag, York, and Mollie's (BYM) model and shared component model (SCM) were used to compare the spatial variations of the relative risk of lung cancer mortality by applying OpenBUGS version 3.2.3 and R version 3.6.1. Results: The median age for death due to lung cancer in Iran is 74 years. During 2011-2014, the age-standardized lung cancer mortality rates among men and women were 12 and 5 per 100,000 individuals, respectively. In addition, almost similar spatial patterns were observed for both men and women. Further, risk factors, which are shared between men and women, were considered as the main cause of variation of lung cancer mortality relative risk in the regions under study for both men and women. The highest impact of the women-specific risk factors was estimated in northeastern and southwestern of the country while the lowest was related to Gilan province in northern part of Iran. Conclusion: Based on the spatial pattern, lung cancer risk factors are at relatively high levels in most parts of Iran, especially in the northwest of the country. Regarding the women, the high-risk regions were considerably extended. Further, the highest concentration of the specific risk factors among women was observed in the eastern, central, and southwestern parts. The smoking effect, and the second-smoking effect and environmental pollutions could play more significant roles for men and women, respectively.
Aesthetic plastic surgery, Jan 3, 2017
Pectus excavatum (PE) is one of the most frequent thoracic malformations. Generally, the malforma... more Pectus excavatum (PE) is one of the most frequent thoracic malformations. Generally, the malformation is not associated with functional disorders and often constitutes an aesthetic alteration with significant psychological distress. To reduce the visibility of the residual scarring produced by corrective surgery and to improve the aesthetic outcome, the authors propose a new prosthetic implant technique through a periareolar access. From January 2005 to January 2015, 11 patients affected with PE underwent the surgical procedure with a sternal prosthesis implanted through a periareolar access with the help of a fiberscope. The preoperative evaluation of the perception of the malformation and postoperation results were made using different questionnaires. The data collected in our series were compared with that reported in 4 different papers where other forms of access were used: sternal, inframammary and transumbilical. No major complications or dislocation of the implants were repor...
In vivo (Athens, Greece), Jan 2, 2017
Evaluation of long-term results after aponeurotic blepharoptosis correction with external levator... more Evaluation of long-term results after aponeurotic blepharoptosis correction with external levator muscle complex advancement. We carried out a retrospective study with medical record review of 20 patients (40 eyes) affected by bilateral aponeurotic moderate and severe ptosis who underwent primary surgery between January 2010 and December 2013. Criteria for outcome evaluations included 3-year postoperative follow-up of upper margin reflex index (uMRD) and symmetry. 3-Year postoperative follow-up showed 17 (85%) cases of successful correction of ptosis and three cases (15%) showed partial success. Two eyes showed hypocorrection, while one eye was overcorrected. The symmetry was maintained in all patients except for the oldest. External superior levator advancement is an effective procedure for moderate and severe aponeurotic blepharoptosis correction, and establishes good long-term eyelid position and symmetry.
Journal of Plastic, Reconstructive & Aesthetic Surgery, 2017
La pubblicazione è resa disponibile sotto le norme e i termini della licenza di deposito, secondo... more La pubblicazione è resa disponibile sotto le norme e i termini della licenza di deposito, secondo quanto stabilito dalla Policy per l'accesso aperto dell'Università degli Studi di Firenze (https://www.sba.unifi.it/upload/policy-oa-2016-1.pdf)
Viene analizzato il legame tra condizione socio-economica e mortalità per tumore al polmone nella... more Viene analizzato il legame tra condizione socio-economica e mortalità per tumore al polmone nella popolazione maschile della Toscana, a livello comunale, nel periodo 1971-94, facendo uso di un modello spazio-temporale Bayesiano con covariate tempo dipendenti. In particolare si intende indagare se tra la manifestazione di un certo livello di deprivazione materiale e quella della mortalità vi sia o meno un tempo di latenza, quindi se la mortalitàè legata alla deprivazione materiale di dieci anni prima (esposizione a fattori di rischio), oppure a quella presente (problemi di accesso al sistema sanitario).
D ip a r tim e nto d i S ta tis tic a " G. Pa r e nti"-V ia le Mo r g a g ni 5 9-5 0 1 3 4 F ir e... more D ip a r tim e nto d i S ta tis tic a " G. Pa r e nti"-V ia le Mo r g a g ni 5 9-5 0 1 3 4 F ir e nz e-w w w .d s .unif i.it
Riassunto: Analizzare congiuntamente la distribuzione spazio-temporale di una malattia in due gru... more Riassunto: Analizzare congiuntamente la distribuzione spazio-temporale di una malattia in due gruppi disgiunti di popolazione permette di fare inferenza su eventuali andamenti comuni ed effettuare ipotesi sulla condivisione di fattori eziologici. In questo lavoro si analizza la mortalità per tumore al polmone nei Comuni della Toscana per gli uomini e le donne appartenenti alle coorti di nati dal 1905 al 1940. Si utilizza una serie di modelli Bayesiani gerarchici spaziotemporali con componenti di varianza comuni e specifiche per genere.
A new approach to ecological regression for disease mapping is introduced, based on semiparametri... more A new approach to ecological regression for disease mapping is introduced, based on semiparametric M-quantile regression models. In particular, we define a Negative Binomial M-quantile model as an alternative to Empirical Bayes or fully Bayesian approaches to disease mapping. The area-level covariates used in ecological regression are usually measured with error, and the proposed M-quantile modelling approach is easily made robust against outlying data in the model covariates. Differences between the M-quantile model and the usual random effects models are discussed, and these alternative approaches are compared using the well-known Scottish Lip cancer data and a simulation experiment. The Lip Cancer data example shows that the Negative Binomial M-quantile model confirms results obtained by other methods, but also seems to have less shrinkage than the Empirical Bayes method, so reducing the problem of oversmoothing. The simulation experiment suggests that the new model leads to estimates with smaller mean square error. We also show how the Negative Binomial M-quantile can be extended to account for spatial correlation between areas using a Geographically Weighted Regression strategy.
Studies in Classification, Data Analysis, and Knowledge Organization, 2012
Two models for jointly analysing the spatial variation of incidences of three (or more) diseases,... more Two models for jointly analysing the spatial variation of incidences of three (or more) diseases, with common and uncommon risk factors, are compared via a simulation experiment. In both models, the linear predictor can be decomposed into shared and disease-specific spatial variability components (named shared clustering and specific clustering respectively). The two models are the shared model on the original formulation that use exchangeable Poisson distribution as response multivariate variable and shared components model that use a Multinomial one. The simulation study shows that models behave similarly. However, Multinomial shared components model performs better for disease-specific spatial variability clustering terms but it is lower for the shared one.
Encyclopedia of Quality of Life and Well-Being Research, 2014
Statistical Modelling, 2001
Statistical Methods in Medical Research, 2014
A new semiparametric approach to model-based small area prediction for counts is proposed and use... more A new semiparametric approach to model-based small area prediction for counts is proposed and used for estimating the average number of visits to physicians for Health Districts in Central Italy. The proposed small area predictor can be viewed as an outlier robust alternative to the more commonly used empirical plug-in predictor that is based on a Poisson generalized linear mixed model with Gaussian random effects. Results from the real data application and from a simulation experiment confirm that the proposed small area predictor has good robustness properties and in some cases can be more efficient than alternative small area approaches.
M-Quantile spatially structured model. The proposed approach is easily made robust against outlyi... more M-Quantile spatially structured model. The proposed approach is easily made robust against outlying data values for covariates. Robust ecological disease mapping is desirable since covariates at area level usually present measure-type error. We
Environmetrics, 2003
ABSTRACT
Environmetrics, 2003
Joint modelling of space and time variation of the risk of disease is an important topic in descr... more Joint modelling of space and time variation of the risk of disease is an important topic in descriptive epidemiology. Most of the proposals in this field deal with at most two time scales (age–period or age–cohort). We propose a hierarchical Bayesian model that can be used as a general framework to jointly study the evolution in time and the spatial pattern of the risk of disease. The rates are modelled as a function of purely spatial terms (local effects of risk factors that do not vary in time), time effects (on the three time axes: age, calendar period and birth cohort) and space–time interactions that describe area specific time patterns. Copyright © 2003 John Wiley & Sons, Ltd.
Environmental and Ecological Statistics, 2005
In this work we propose a Bayesian ecological analysis in which a latent variable summarizes data... more In this work we propose a Bayesian ecological analysis in which a latent variable summarizes data on emissions of atmospheric pollutants. We specified a hierarchical Bayesian model with spatially structured and unstructured random terms with a nested latent factor model. This can be considered a combination of the convolution spatial model of Besag et al. (1991) and an ecological regression analysis in which a latent variable plays the role of the covariate. The unified approach allows to proper account for the uncertainty in the latent score estimation in the regression analysis. The Bayesian Latent Factor model is used to summarize the information on environmental pressure derived from three stressors: Carbon Monoxide, Nitrogen Oxides and Inhalable Particles. We found evidence of positive correlation between Lung cancer mortality and environmental pressure indicators, in males, Tuscany (Italy), 1995-1999. Environmental pressure seems to be restricted to fourteen municipalities (top 5% of the Latent Factor distribution). The model identified two areas with high point source emissions.