Chronic diseases, as per the World Health Organization (WHO), are characterized by long duration, slow progression, and arise from a combination of genetic, physiological, environmental, and behavioral factors. Globally, they account for 74% of all deaths, totaling 41 million deaths annually. Managing chronic diseases necessitates regular medication intake throughout an individual's lifetime, underscoring the significance of medication adherence especially in an aging world that is constantly more prone to chronic diseases. Numerous digital solutions in healthcare already exist to aid chronic patients in managing their medication adherence. However, these solutions exhibit various limitations, particularly in the personalization of interventions tailored to different types of end-users. Thus, the general aim and focus of this thesis is to develop a digital health framework to improve medication adherence and quality of life in chronic patients, with a focus on hypertension and chronic cardiac conditions, by leveraging user-centered design (UCD) methods, such as Personas, to tailor personalized persuasive system design (PSD) techniques to increase the engagement and efficacy of such framework. In Chapter 1 of this dissertation, an introduction to the contextual background of the study is presented. The state of the art in the management of medication adherence through digital solutions is provided and discussed. The main topics relevant to UCD and PSD are presented and explained. Focus is put on the concept of Personas, their use and implementation within current solutions, as well as the relevant state of the art regarding their development and limitations. Finally, the general research objective as well as the specific research questions that have driven the research effort of this thesis are presented in detail. In Chapter 2, a novel framework for the development of quantitative Personas is presented and tested on two different datasets: the former regarding burnout in nurses due to COVID-19, and the latter regarding the willingness of vaccination against COVID-19 in the general population. The resulting Personas from the application of this method are presented and discussed, together with the novelties compared to current state of the art methods for their development. The proposed framework aims at supporting the identification of the common characteristics within a given population, allowing to identify the needs and requirements of different clusters, facilitating the development of tailored interventions. One of the main limitations of Personas is the lack, in current scientific literature, of methods for their validation. For this reason, in Chapter 3 a novel method for Personas validation is proposed and tested on the previously described datasets. This novel method leverages Self-Supervised Machine Learning techniques to train a model that is capable of discerning between the developed Personas, assessing their capability of capturing the most important features of the target population. The validation of the developed Personas further confirms their capability in supporting researchers in applying PSD techniques. In Chapter 4, the InTakeCare platform is presented and discussed. This platform represents the novel digital health framework that can support chronic patients in managing their medication adherence. It uses a three-tier architecture structure that enables modularity and scalability of the solution, involving both patients and physicians. A set of modules involving a Skill for Amazon Alexa devices, a Web Dashboard for patients and physicians, a Mobile application and a Skill for the Furhat social robot are presented in detail and discussed. The usability studies of the Amazon Alexa Skill for patients and the Web Dashboard for physicians are presented and discussed in Chapter 5. The protocols are presented in detail, and the results highlight the usability of the systems and their unobtrusiveness in the everyday life of patients and physicians respectively. These studies confirm the capabilities of the InTakeCare platform, while suggesting new features and future works. In Chapter 6 the InTakeCare study is presented in its entirety. It represents the combination of the Persona development, validation, and implementation of the presented platform. Personas for medication adherence in chronic patients are developed and validated starting from focus groups, using the previously presented methods. Lastly, in Chapter 7 the findings of this thesis are further extended and discussed. The novelties introduced are correlated to the general research question and to the specific research objective of this thesis, highlighting the strengths of Personas creation and validation in supporting the development of tailored interventions that can increase medication adherence in chronic patients. Furthermore, the limitations of these studies are highlighted and discussed, and future wors and research lines are proposed.
Le malattie croniche, o non trasmissibili, secondo l'Organizzazione Mondiale della Sanità (OMS), sono caratterizzate da una lunga durata, da una lenta progressione e da una combinazione di fattori genetici, fisiologici, ambientali e comportamentali. A livello globale, sono responsabili del 74% di tutti i decessi, per un totale di 41 milioni di morti all'anno. La gestione delle malattie croniche richiede l'assunzione regolare di farmaci per tutta la vita dell'individuo, sottolineando l'importanza dell'aderenza ai farmaci soprattutto in un mondo che invecchia e che è sempre più soggetto a malattie croniche. Esistono già numerose soluzioni digitali in ambito sanitario per aiutare i pazienti cronici a gestire l'aderenza ai farmaci. Tuttavia, queste soluzioni presentano diversi limiti, in particolare nella personalizzazione degli interventi su misura per i diversi tipi di utenti finali. Pertanto, l'obiettivo generale di questa tesi è quello di sviluppare un framework di salute digitale per migliorare l'aderenza ai farmaci e la qualità della vita nei pazienti cronici, con particolare attenzione all'ipertensione e alle patologie cardiache croniche, sfruttando metodi di progettazione incentrati sull'utente (User-Centered Design, UCD), come le Personas, per adattare tecniche di design di sistemi persuasivi (Persuasive System Design, PSD) personalizzate per aumentare il coinvolgimento e l'efficacia di tale framework. Nel Capitolo 1 di questa tesi viene presentata un'introduzione al contesto dello studio. Viene fornito e discusso lo stato dell'arte della gestione dell'aderenza ai farmaci attraverso soluzioni digitali. Vengono presentati e spiegati i principali argomenti rilevanti per UCD e PSD. L'attenzione viene posta sul concetto di Personas, sul loro utilizzo e implementazione all'interno delle soluzioni attuali, nonché sullo stato dell'arte relativo al loro sviluppo e ai loro limiti. Infine, vengono presentati in dettaglio l'obiettivo generale della ricerca e le domande specifiche che hanno guidato lo sforzo di ricerca di questa tesi. Nel Capitolo 2, viene presentato un nuovo framework per lo sviluppo di Personas quantitativi, testato su due diversi set di dati: il primo riguardante il burnout degli infermieri dovuto alla COVID-19 e il secondo riguardante la disponibilità a vaccinarsi contro la COVID-19 nella popolazione generale. Le Personas risultanti dall'applicazione di questo metodo sono presentate e discusse, insieme alle novità rispetto agli attuali metodi allo stato dell'arte per il loro sviluppo. Il framework proposto mira a supportare l'identificazione delle caratteristiche comuni all'interno di una determinata popolazione, consentendo di individuare i bisogni e le esigenze dei diversi cluster, facilitando lo sviluppo di interventi su misura. Uno dei principali limiti delle Personas è la mancanza, nell'attuale letteratura scientifica, di metodi per la loro validazione. Per questo motivo, nel Capitolo 3 viene proposto un metodo innovativo per la validazione delle Personas, testato sui set di dati precedentemente descritti. Questo metodo innovativo sfrutta tecniche di apprendimento automatico auto-supervisionato (Self-Supervised Machine Learning) per addestrare un modello in grado di discernere tra le Personas sviluppate, valutando la loro capacità di catturare le caratteristiche più importanti della popolazione target. La validazione delle Personas sviluppate conferma ulteriormente la loro capacità di supportare i ricercatori nell'applicazione delle tecniche di PSD. Nel Capitolo 4 viene presentata e discussa la piattaforma InTakeCare. Questa piattaforma rappresenta un nuovo framework di salute digitale in grado di supportare i pazienti cronici nella gestione dell'aderenza ai farmaci. Utilizza una struttura architettonica a tre livelli che consente la modularità e la scalabilità della soluzione, coinvolgendo sia i pazienti che i medici. Una serie di moduli che comprendono una Skill per i dispositivi Amazon Alexa, una Dashboard Web per pazienti e medici, un'applicazione mobile e una Skill per il robot sociale Furhat sono presentati in dettaglio e discussi. Gli studi di usabilità della Skill Amazon Alexa per i pazienti e della Dashboard Web per i medici sono presentati e discussi nel Capitolo 5. I protocolli sono presentati in dettaglio e i risultati evidenziano l'usabilità dei sistemi e la loro discreta presenza nella vita quotidiana rispettivamente dei pazienti e dei medici. Questi studi confermano le capacità della piattaforma InTakeCare, suggerendo al contempo nuove funzionalità e lavori futuri. Nel Capitolo 6 viene presentato lo studio InTakeCare nella sua interezza. Rappresenta la combinazione dello sviluppo di Persona, della validazione e dell'implementazione della piattaforma presentata. Le Persona per l'aderenza ai farmaci nei pazienti cronici vengono sviluppate e validate a partire dai focus group, utilizzando i metodi precedentemente presentati. Infine, nel Capitolo 7 i risultati di questa tesi vengono ulteriormente estesi e discussi. Le novità introdotte sono correlate alla domanda di ricerca generale e all'obiettivo di ricerca specifico di questa tesi, evidenziando i punti di forza della creazione e della validazione di Personas nel supportare lo sviluppo di interventi personalizzati che possono aumentare l'aderenza ai farmaci nei pazienti cronici. Inoltre, vengono evidenziati e discussi i limiti di questi studi e si propongono futuri sviluppi e linee di ricerca.
Development of a framework for chronic cardiologic patient engagement towards improved wellbeing using persuasive technology design approaches
TAURO, EMANUELE
2023/2024
Abstract
Chronic diseases, as per the World Health Organization (WHO), are characterized by long duration, slow progression, and arise from a combination of genetic, physiological, environmental, and behavioral factors. Globally, they account for 74% of all deaths, totaling 41 million deaths annually. Managing chronic diseases necessitates regular medication intake throughout an individual's lifetime, underscoring the significance of medication adherence especially in an aging world that is constantly more prone to chronic diseases. Numerous digital solutions in healthcare already exist to aid chronic patients in managing their medication adherence. However, these solutions exhibit various limitations, particularly in the personalization of interventions tailored to different types of end-users. Thus, the general aim and focus of this thesis is to develop a digital health framework to improve medication adherence and quality of life in chronic patients, with a focus on hypertension and chronic cardiac conditions, by leveraging user-centered design (UCD) methods, such as Personas, to tailor personalized persuasive system design (PSD) techniques to increase the engagement and efficacy of such framework. In Chapter 1 of this dissertation, an introduction to the contextual background of the study is presented. The state of the art in the management of medication adherence through digital solutions is provided and discussed. The main topics relevant to UCD and PSD are presented and explained. Focus is put on the concept of Personas, their use and implementation within current solutions, as well as the relevant state of the art regarding their development and limitations. Finally, the general research objective as well as the specific research questions that have driven the research effort of this thesis are presented in detail. In Chapter 2, a novel framework for the development of quantitative Personas is presented and tested on two different datasets: the former regarding burnout in nurses due to COVID-19, and the latter regarding the willingness of vaccination against COVID-19 in the general population. The resulting Personas from the application of this method are presented and discussed, together with the novelties compared to current state of the art methods for their development. The proposed framework aims at supporting the identification of the common characteristics within a given population, allowing to identify the needs and requirements of different clusters, facilitating the development of tailored interventions. One of the main limitations of Personas is the lack, in current scientific literature, of methods for their validation. For this reason, in Chapter 3 a novel method for Personas validation is proposed and tested on the previously described datasets. This novel method leverages Self-Supervised Machine Learning techniques to train a model that is capable of discerning between the developed Personas, assessing their capability of capturing the most important features of the target population. The validation of the developed Personas further confirms their capability in supporting researchers in applying PSD techniques. In Chapter 4, the InTakeCare platform is presented and discussed. This platform represents the novel digital health framework that can support chronic patients in managing their medication adherence. It uses a three-tier architecture structure that enables modularity and scalability of the solution, involving both patients and physicians. A set of modules involving a Skill for Amazon Alexa devices, a Web Dashboard for patients and physicians, a Mobile application and a Skill for the Furhat social robot are presented in detail and discussed. The usability studies of the Amazon Alexa Skill for patients and the Web Dashboard for physicians are presented and discussed in Chapter 5. The protocols are presented in detail, and the results highlight the usability of the systems and their unobtrusiveness in the everyday life of patients and physicians respectively. These studies confirm the capabilities of the InTakeCare platform, while suggesting new features and future works. In Chapter 6 the InTakeCare study is presented in its entirety. It represents the combination of the Persona development, validation, and implementation of the presented platform. Personas for medication adherence in chronic patients are developed and validated starting from focus groups, using the previously presented methods. Lastly, in Chapter 7 the findings of this thesis are further extended and discussed. The novelties introduced are correlated to the general research question and to the specific research objective of this thesis, highlighting the strengths of Personas creation and validation in supporting the development of tailored interventions that can increase medication adherence in chronic patients. Furthermore, the limitations of these studies are highlighted and discussed, and future wors and research lines are proposed.File | Dimensione | Formato | |
---|---|---|---|
Tesi_Dottorato_Tauro.pdf
solo utenti autorizzati a partire dal 03/04/2027
Dimensione
8.68 MB
Formato
Adobe PDF
|
8.68 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/220413