In this thesis, we model the life-cycle evolution of individual healthcare expenditures, expressed as a function of the aging process, health shocks and conditions, and distance to death. All the analyses are carried out by using a unique dataset, which allows us to focus on different types of healthcare services and different subsamples of individuals. The population of interest consists of individuals aged 50-70, the age window where the first adverse health events are expected to arise. In the first chapter, we use a two-way fixed effects model to examine the effect of age, morbidity, and time to death (TTD) on individual healthcare expenditures (HCE). The estimation is carried out by controlling for several confounding factors, including individual and General Practitioner (GP) fixed effects. We also investigate to what extent patients’ and GP’ characteristics contribute to the overall variability in expenditures among individuals. Our main results show that age, morbidity, and TTD are all important determinants of HCE and are among the elements that contribute most to the variability in HCE among individuals. Total HCE is increasing in age, with the latter found to be negatively correlated with the time to death, a result in contrast with the ‘red herring’ hypothesis. Such an increase with age of overall expenditures is mainly driven by expenses for out-of-hospital services; in contrast, no difference in hospital costs is observed over the considered lifespan once the other factors are taken into account. On the other hand, inpatient expenditures mainly drive the morbidity and end-of-life profiles of total HCE. Concerning heterogeneous analysis, we find that chronic and disabled individuals with health shocks requiring hospitalization are those who place the greatest burden on the costs borne by the Italian healthcare system. It suggests that the enhancement of preventive approaches before the onset of such shocks is a priority goal to reduce the incidence of long-lasting diseases and prevent them from deteriorating to the point of exacerbation in acute cases requiring hospital admissions. Given the results obtained in the first chapter, in the second one, we use a difference-in-difference event study approach to estimate the short- and long-run impact of the hospitalization on HCE, with hospital admissions analyzed here as a measurable subset of those first adverse health events individuals aged 50-70 experience in their life. Our main findings confirm the existence of a large effect of the first hospitalization on HCE and show that the first access is associated with substantial future medical expenses in all healthcare settings, accounted for the largest part by acute inpatient care. Indeed, the analysis of hospital expenditures indicates the occurrence of subsequent hospitalizations, mainly required for complications of cardiovascular diseases and cancer. The latter are responsible for the highest increase in inpatient expenditures and present a persistent post-admission increase also in outpatient and pharmaceutical expenses, a result driven by the high incidence of chronic and disabled individuals within the group of those affected by these two conditions. From a policy perspective, it indicates need for a strengthening of territorial care and tertiary prevention improvements, necessary to soften the impact of ongoing illnesses with lasting effects. On the one hand, it would improve patients’ health by preventing complications and acute cases; on the other hand, it would also generate significant savings through reduced avoidable additional hospitalizations.

In questa tesi modelliamo l'evoluzione nel ciclo di vita delle spese sanitarie individuali (HCE), espresse in funzione del processo di invecchiamento, degli shock e delle condizioni di salute e della distanza dalla morte. Tutte le analisi sono condotte utilizzando un unico dataset, che permette di analizzare diversi tipi di servizi sanitari e diversi sottocampioni di individui. La popolazione di interesse è costituita da individui di età compresa tra i 50 e i 70 anni, la finestra di età in cui si verificano i primi eventi sanitari avversi. Nel primo capitolo, usiamo un two-way fixed effects model per esaminare l'effetto dell'età, della morbilità e della distanza al decesso (TTD) sulle spese sanitarie individuali. La stima viene effettuata includendo diversi controlli, compresi gli effetti fissi individuali e del medico di medicina generale. Indaghiamo anche in che misura le caratteristiche dei pazienti e dei medici di base contribuiscono alla variabilità complessiva delle spese tra gli individui. I nostri risultati principali mostrano che l'età, la morbilità e TTD sono tutti importanti determinanti della spesa e sono tra gli elementi che contribuiscono maggiormente alla variabilità della stessa tra gli individui. La spesa totale aumenta con l'età, e quest'ultima risulta essere correlata negativamente con il tempo alla morte, un risultato in contrasto con la “red herring” hypothesis. Tale aumento con l'età delle spese complessive è principalmente guidato dalle spese per i servizi extra-ospedalieri; al contrario, non si osserva alcuna differenza nei costi ospedalieri durante la durata l’arco di vita considerato, una volta presi in considerazione gli altri fattori. D'altra parte, le spese ospedaliere guidano principalmente i profili di morbilità e di fine vita della spesa totale. Per quanto riguarda l'analisi di eterogeneità, gli individui cronici e disabili con shock sanitari che richiedono l'ospedalizzazione sono quelli che gravano maggiormente sui costi sostenuti dal sistema sanitario italiano. Ciò suggerisce che il potenziamento degli approcci preventivi prima dell'insorgenza di tali shock è un obiettivo di policy prioritario per ridurre l'incidenza delle malattie di lunga durata ed evitare che queste si aggravino fino a trasformarsi in casi acuti che richiedono ricoveri ospedalieri. Dati i risultati ottenuti nel primo capitolo, nel secondo usiamo un difference-in-difference event study per stimare l'impatto a breve e lungo termine dell'ospedalizzazione sulla spesa sanitaria individuale. In questo capitolo, i ricoveri ospedalieri sono analizzati come sottoinsieme misurabile di quei primi eventi sanitari avversi che gli individui di 50-70 anni sperimentano nella loro vita. I nostri risultati principali confermano l'esistenza di un effetto rilevante del primo ricovero sulla spesa e mostrano che il primo accesso è associato a sostanziali spese mediche future in tutti i contesti sanitari, e soprattutto nel settore delle cure ospedaliere acute. L'analisi delle spese ospedaliere infatti indica il verificarsi di successivi ricoveri, richiesti principalmente per complicazioni di malattie cardiovascolari e cancro. Queste patologie sono responsabili del maggiore aumento delle spese di ricovero e presentano un persistente aumento post-ricovero anche nelle spese ambulatoriali e farmaceutiche, un risultato guidato dall'alta incidenza di individui cronici e disabili all'interno del gruppo degli individui affetti da queste due condizioni. Da un punto di vista di policy, questo risultato indica la necessità di un rafforzamento dell'assistenza territoriale e di miglioramenti nella prevenzione terziaria, necessari per attenuare l'impatto delle malattie croniche e di lunga durata in corso. Da un lato, questo migliorerebbe la salute dei pazienti prevenendo complicazioni e casi acuti; dall'altro, genererebbe un risparmio significativo attraverso la riduzione di ricoveri evitabili.

(2021). Healthcare Expenditures for the Young-Old Population. (Tesi di dottorato, Università degli Studi di Milano-Bicocca, 2021).

Healthcare Expenditures for the Young-Old Population

TORRINI, IRENE
2021

Abstract

In this thesis, we model the life-cycle evolution of individual healthcare expenditures, expressed as a function of the aging process, health shocks and conditions, and distance to death. All the analyses are carried out by using a unique dataset, which allows us to focus on different types of healthcare services and different subsamples of individuals. The population of interest consists of individuals aged 50-70, the age window where the first adverse health events are expected to arise. In the first chapter, we use a two-way fixed effects model to examine the effect of age, morbidity, and time to death (TTD) on individual healthcare expenditures (HCE). The estimation is carried out by controlling for several confounding factors, including individual and General Practitioner (GP) fixed effects. We also investigate to what extent patients’ and GP’ characteristics contribute to the overall variability in expenditures among individuals. Our main results show that age, morbidity, and TTD are all important determinants of HCE and are among the elements that contribute most to the variability in HCE among individuals. Total HCE is increasing in age, with the latter found to be negatively correlated with the time to death, a result in contrast with the ‘red herring’ hypothesis. Such an increase with age of overall expenditures is mainly driven by expenses for out-of-hospital services; in contrast, no difference in hospital costs is observed over the considered lifespan once the other factors are taken into account. On the other hand, inpatient expenditures mainly drive the morbidity and end-of-life profiles of total HCE. Concerning heterogeneous analysis, we find that chronic and disabled individuals with health shocks requiring hospitalization are those who place the greatest burden on the costs borne by the Italian healthcare system. It suggests that the enhancement of preventive approaches before the onset of such shocks is a priority goal to reduce the incidence of long-lasting diseases and prevent them from deteriorating to the point of exacerbation in acute cases requiring hospital admissions. Given the results obtained in the first chapter, in the second one, we use a difference-in-difference event study approach to estimate the short- and long-run impact of the hospitalization on HCE, with hospital admissions analyzed here as a measurable subset of those first adverse health events individuals aged 50-70 experience in their life. Our main findings confirm the existence of a large effect of the first hospitalization on HCE and show that the first access is associated with substantial future medical expenses in all healthcare settings, accounted for the largest part by acute inpatient care. Indeed, the analysis of hospital expenditures indicates the occurrence of subsequent hospitalizations, mainly required for complications of cardiovascular diseases and cancer. The latter are responsible for the highest increase in inpatient expenditures and present a persistent post-admission increase also in outpatient and pharmaceutical expenses, a result driven by the high incidence of chronic and disabled individuals within the group of those affected by these two conditions. From a policy perspective, it indicates need for a strengthening of territorial care and tertiary prevention improvements, necessary to soften the impact of ongoing illnesses with lasting effects. On the one hand, it would improve patients’ health by preventing complications and acute cases; on the other hand, it would also generate significant savings through reduced avoidable additional hospitalizations.
CELLA, MICHELA
LUCIFORA, CLAUDIO
Salute; Cure Sanitarie; Spesa sanitaria; Determinanti; Medico di Base
Health; Healthcare; Health Expenditures; Determinants; Medico di Base
SECS-P/01 - ECONOMIA POLITICA
English
23-set-2021
ECONOMIA - DEFAP
33
2019/2020
open
(2021). Healthcare Expenditures for the Young-Old Population. (Tesi di dottorato, Università degli Studi di Milano-Bicocca, 2021).
File in questo prodotto:
File Dimensione Formato  
phd_unimib_833677.pdf

accesso aperto

Descrizione: Tesi di "Torrini" - "Irene" - "833677"
Tipologia di allegato: Doctoral thesis
Dimensione 3.11 MB
Formato Adobe PDF
3.11 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/330206
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
Social impact