Daniele Bellavia

A critical aspect of modern healthcare involves recognizing and addressing pharmaceutical needs. Predictive models serve as valuable decision-making tools in the healthcare sector to proactively prevent supply chain failures. However, training these models on real historical data to reliably reflect actual demand is a delicate process. An effective model, capable of estimating the amount of drugs to be distributed in relation to the patient’s needs, must be accurate and inherently fair. Our study endeavors to bridge legal perspectives on fairness with practical assessments of algorithmic fairness, specifically in the context of predicting drugs to be distributed in a specific studied area of reference. An in-depth overview of the Italian National Healthcare Service is provided, emphasizing its regulatory role in drug dispensation and its inherent challenges. Furthermore, a review of fundamental bias research principles is provided, encompassing legal and statistical viewpoints. In addition, a comprehensive Exploratory Data Analysis is conducted using real world data, to highlight challenges that can be encountered in the initial modeling phase. The results of the analysis reveal the presence of missing values in some of the most relevant fields, and signal differences in the drugs distribution patterns between the two genders for specific therapeutic groups. Such disparities would require further investigations to verify the presence of social bias. These findings contribute to an in-depth understanding of patient populations concerning drug collection. Importantly, our study promotes a comprehensive approach that incorporates legal considerations and technical elements to improve the fairness and efficacy of predictive models in healthcare.

Background and importance: Automation of hospital medication management demonstrated advantages to wards manual systems, especially in error reduction, improving patient safety and ensuring drugs’ traceability. Despite the existence of literature on benefits, no multidimensional evidence on automation of hospital medication management is available. Aim and objectives: The study aimed to demonstrate the value of four scenarios of automated technologies’ introduction, with a comprehensive health technology assessment (HTA) approach, comparing: (1) manual dispensing, (2) presence of only centralised automated systems in the hospital pharmacy, (3) presence of only decentralised automated systems in the wards and (4) integration of scenarios 2 and 3 into a full solution, with electronic prescription. Material and methods: The HTA involved 50 healthcare professionals (pharmacists, nurses, decision-makers and other professionals) in four European countries in 2021. After a structured literature review, the nine domains of the EunetHTA Core Model were deployed using validated questionnaires (with a seven-item Likert scale). Differences among groups and scenarios were studied by ANOVA test. All analyses were conducted considering a level of significance equal to 0.05 and were performed using IBM SPSS software (Version 22.0). Results: Results from the efficacy and safety questionnaires showed that the presence of automation resulted in a decrease in dispensing errors (1.75, 1.20, 1.88, 2.19, respectively, for scenarios 1, 2, 3, 4; p value = 0.000) and consequently in adverse events (–2.13, 1.18, 1.71, 2.46, respectively, for scenarios 1, 2, 3, 4; p value = 0.000), especially if associated with electronic prescribing, confirming the literature findings. A low organisational impact of automation was registered (–0.71, 0.50, 0.49, 0.63, respectively, for scenarios 1, 2, 3, 4) due to a trade-off between technological change efforts and efficiency beneficial effects in the first year. Ethical and social dimension results demonstrated a positive impact of automation (–0.93, 0.72, 1.03, 1.23, respectively for scenarios 1, 2, 3, 4; p value = 0.000) on patients’ perceived quality of life. The impact on drugs thefts and the identification of responsibility in cases of legal controversies were the most appreciated legal items. Conclusion and relevance: In a literature dominated by safety evidence on automated solutions, a complete HTA approach demonstrates its validity in communicating and demonstrating multidimensional and multidisciplinary values of hospital automated dispensing solutions.