Development and validation of a risk model for prediction of hazardous alcohol consumption in general practice attendees : the PredictAL study

Background: Little is known about the risk of progression to hazardous alcohol use in people currently drinking at safe limits. We aimed to develop a prediction model (predictAL) for the development of hazardous drinking in safe drinkers. Methods: A prospective cohort study of adult general practice attendees in six European countries and Chile followed up over 6 months. We recruited 10,045 attendees between April 2003 to February 2005. 6193 European and 2462 Chilean attendees recorded AUDIT scores below 8 in men and 5 in women at recruitment and were used in modelling risk. 38 risk factors were measured to construct a risk model for the development of hazardous drinking using stepwise logistic regression. The model was corrected for over fitting and tested in an external population. The main outcome was hazardous drinking defined by an AUDIT score >= 8 in men and >= 5 in women. Results: 69.0% of attendees were recruited, of whom 89.5% participated again after six months. The risk factors in the final predictAL model were sex, age, country, baseline AUDIT score, panic syndrome and lifetime alcohol problem. The predictAL model's average c-index across all six European countries was 0.839 (95% CI 0.805, 0.873). The Hedge's g effect size for the difference in log odds of predicted probability between safe drinkers in Europe who subsequently developed hazardous alcohol use and those who did not was 1.38 (95% CI 1.25, 1.51). External validation of the algorithm in Chilean safe drinkers resulted in a c-index of 0.781 (95% CI 0.717, 0.846) and Hedge's g of 0.68 (95% CI 0.57, 0.78). Conclusions: The predictAL risk model for development of hazardous consumption in safe drinkers compares favourably with risk algorithms for disorders in other medical settings and can be a useful first step in prevention of alcohol misuse.

Evaluation of long term persistence of Diabetes Mellitus Type II treatment newly diagnosed patients in Lisbon and Tagus health administration region : a three year follow up study

RESUMO: Raional: A persistência à terapêutica é o tempo em qualquer antidiabético oral, desde o seu início até à descontinuação de todas as medicações ou até ao fim do período do estudo. Os objetivos deste estudo foi a análise da persistência à terapêutica no segundo e terceiro anos após início do tratamento em doentes adultos diagnosticados na região de Lisboa e Vale do Tejo e determinar o efeito de determinadas variáveis na persistência a longo prazo. Métodos: Um estudo retrospetivo não interventivo foi desenhado com base nos dados a obter do SIARS (prescrições e aquisições na farmácia) e Pordata. A persistência foi quantificada como a percentagem de doentes que continuam a adquirir pelo menos um antidiabético oral ao segundo e terceiro anos após a compra da primeira receita. A associação entre a persistência e o segundo e terceiro anos com cada uma das co-variáveis foi aferido pelo teste qui-quadrado e os odd ratios foram calculados com recurso a um modelo de regressão logística. Resultados: A persistência à terapêutica obtida foi de 80% e 62% para o segundo e terceiro anos após início da terapêutica. Odd ratios para primeiro e segundo ano: número de grupos farmacoterapêuticos diferentes (OR = 2.167, 1.807 – 2.598, p = 0.000 / OR = 1.863, 1.621 – 2.142, p = 0.000); idade (OR = 0.914, 0.772 – 1.081, p = 0.294 / OR = 0.875, 0.764 – 1.002, p = 0.054); sexo (OR = 1.163, 0.983 – 1.377, p = 0.079); número de diferentes prescritores (OR = 3.594, 3.030 – 4.262, p = 0.000 / OR = 2.167, 1.886 – 2.491, p = 0.000); instituição de prescrição (OR = 0.725, 0.698 – 0.753, p = 0.000 / OR = 0.683, 0.650 – 0.717, p = 0.000); grupo farmacoterapêutico (OR = 1.056, 1.043 – 1.069, p = 0.000 / OR = 1.077, 1.060 – 1.095, p = 0.000); relação com o médico (OR = 0.834, 0.816 – 0.852, p = 0.000 / OR = 0.799, 0.777 – 0.821, p = 0.000) e custo médio mensal por grupo farmacoterapêutico (OR = 0.954, 0.942 – 0.968, p = 0.000 / OR = 0.930, 0.914 – 0.947, p = 0.000). Conclusões: O valor da persistência à terapêutica no segundo ano é ligeiramente acima do que é mencionado na literatura e não existem dados para comparar os resultados do terceiro ano. Relativamente ao efeito das co-variáveis no segundo e terceiro anos após o início do tratamento, os resultados são sobreponíveis, sendo que o sexo não está associado à persistência ao terceiro ano.----------------------------------ABSTRACT: Background: Therapy persistence is the time on any antidiabetic medication, from initiation of therapy to discontinuation of all medications or the end of the study period. The aim of the study was to analyse the therapy persistence in the second and third years after treatment initiation in newly diagnosed adult patients in the Lisbon and Tagus Valley region and to determine the effect of several co-variables in the long term persistence. Methods: A retrospective non-interventional study based on SIARS data (drug prescriptions and acquisitions) and Pordata was designed. Persistence was quantified as the percentage of patients that continued to purchase at least one type of antidiabetic at year 2 and 3 after the date of first prescription acquisition. Association between persistence at second and third years with each of the other co-variables were verified by using the Chi-Square test and odds ratio were calculated using a regression logistic model. Results: Therapy persistence obtained was 80% and 62% for the second and third years after treatment initiation. Odd ratios for second and third years: number of different pharmacotherapeutic groups (OR = 2.167, 1.807 – 2.598, p = 0.000 / OR = 1.863, 1.621 – 2.142, p = 0.000); age (OR = 0.914, 0.772 – 1.081, p = 0.294 / OR = 0.875, 0.764 – 1.002, p = 0.054); gender (OR = 1.163, 0.983 – 1.377, p = 0.079); number of different prescribers (OR = 3.594, 3.030 – 4.262, p = 0.000 / OR = 2.167, 1.886 – 2.491, p = 0.000); institution of prescription (OR = 0.725, 0.698 – 0.753, p = 0.000 / OR = 0.683, 0.650 – 0.717, p = 0.000); pharmacotherapeutic group (OR = 1.056, 1.043 – 1.069, p = 0.000 / OR = 1.077, 1.060 – 1.095, p = 0.000); relationship with the physician (OR = 0.834, 0.816 – 0.852, p = 0.000 / OR = 0.799, 0.777 – 0.821, p = 0.000) and average cost per month and per pharmacotherapeutic group (OR = 0.954, 0.942 – 0.968, p = 0.000 / OR = 0.930, 0.914 – 0.947, p = 0.000). Conclusions: Second year therapy persistence value is slightly above of what is referenced in literature and no data was found to compare the third year value. Regarding the effect of the co-variables analysed at second and third years after treatment initiation, the results were overlapping with gender being not associated with persistence at the third year.

Credit default swap (CDS) prediction model & trading strategy

This project focuses on the study of different explanatory models for the behavior of CDS security, such as Fixed-Effect Model, GLS Random-Effect Model, Pooled OLS and Quantile Regression Model. After determining the best fitness model, trading strategies with long and short positions in CDS have been developed. Due to some specifications of CDS, I conclude that the quantile regression is the most efficient model to estimate the data. The P&L and Sharpe Ratio of the strategy are analyzed using a backtesting analogy, where I conclude that, mainly for non-financial companies, the model allows traders to take advantage of and profit from arbitrages.