Gianluca Baio | PubMed Health

PubMed2006

Handling manipulated evidence.

Bayesian Networks have been advocated as useful tools to describe the relations of dependence/independence among random variables and relevant hypotheses in a crime case. Moreover, they have been applied to help the investigator structure the problem and evaluate the impact of the observed evidence, typically with respect to the hypothesis of guilt of a suspect. In this paper we describe a model to handle the possibility that one or more pieces of evidence have been manipulated in order to mislead the investigations. This method is based on causal inference models, although it is developed in a different, specific framework.

Bayes Theorem↗

PubMed2005

Economic evaluation of HIV treatments: the I.CO.N.A. cohort study.

OBJECTIVE: To describe the changes in costs of care for HIV-positive patients in Italy after the spread of antiretroviral combination therapies (HAART). METHODS: Five thousand four hundred and twenty-two patients from the I.CO.N.A. (Italian Cohort Naive Antiretrovirals) study were followed between 1997 and 2002. Costs included antiretroviral therapies (ART), hospital admissions, prophylaxis, and main laboratory examinations. The perspective was that of the National Health Service. RESULTS: Admission costs per person-year decreased from 2148 euro in 1997 to 256 in 2002, while the average annual costs of ART increased from 2145 to 3149 euro (1997 prices). From 1997 to 1999, ART costs increased from 42.3 to 85.9% of the total, while admission costs decreased from 42.3 to 7.0% and prophylaxis from 7.3 to 1.7%. The breakdown of ART costs shows how dual therapies decreased over time in favor of HAART, falling from 26.8% in 1997 to 5.9% in 2002. Patients with fewer than five treatment switches had the lowest costs distributions over the entire observation period. CONCLUSIONS: From 1997 to 2002 inpatient costs progressively decreased in favor of antiretroviral therapy. Annual average costs per patient decreased, while total direct costs increased over time: health resources, initially concentrated on hospitalized patients were then distributed over a growing number of subjects.

Antiretroviral Therapy, Highly Active↗

PubMed2002

A retrospective, population-based analysis of persistence with antihypertensive drug therapy in primary care practice in Italy.

BACKGROUND: Many hypertensive patients discontinue drug therapy despite the increased cardiovascular risk associated with inadequately controlled hypertension. However, most experiments do not address this premature discontinuation of therapy, making it difficult to project the appropriateness of antihypertensive drugs in real-world use. OBJECTIVE: The goal of this study was to assess patients' persistence with antihypertensive drug therapy in a nonexperimental setting. METHODS: An administrative database kept by the Local Health Unit of Ravenna, Ravenna, Italy, listing patient baseline characteristics, drug prescriptions, and hospital admissions was used to perform a population-based, retrospective study. The study included all patients who met the following criteria from January I through December 31, 1997: new user of antihypertensive drugs; > or = 20 years of age; receiving a first prescription for a diuretic, beta-blocker, calcium channel blocker, angiotensin II-receptor antagonist (AIIA), or angiotensin-converting enzyme inhibitor. All prescriptions for antihypertensive drugs filled during the 12-month follow-up period were used to define patients as continuers, switchers, or discontinuers on the basis of their persistence with therapy. RESULTS: A total of 16,783 patients were included in the study analyses: 7,409 men (44.1%) and 9,374 women (55.9%), with an average age of 56.1 +/- 18.3 years (range, 20-105 years). Of this study population, 64.9% (n = 10,894) discontinued treatment over the course of follow-up, 26.9% (n = 4508) continued treatment with the initially prescribed medication (with 5.1% [n = 862] adding another medication for combination therapy), and 8.2% (n = 1381) switched medications. Patients initially prescribed AIIAs were more likely to continue treatment than those initiated on other types of antihypertensives (P < 0.001). Discontinuation was associated with younger age, lower prevalence of concurrent chronic pharmacotherapies, and lower prevalence of previous hospitalizations for cardiovascular disease (all P < 0.001). CONCLUSIONS: Health care claims data are a powerful tool for measuring continuation of therapy, providing detailed, populationwide epidemiologic and economic information for analyzing antihypertensive drug treatment. Further studies are required to relate pharmacotherapy to outcomes.

Adult↗