Introduction Criminal responsibility and the insanity defense represent complex and widely debated issues in forensic psychiatry. The lack of shared scientific standards, together with the potential influence of cognitive biases, may affect expert evaluations, with possible implications for judicial outcomes. The present study aimed to analyze, within an Italian judicial case series, the methodology adopted by expert witnesses and consultants, the level of agreement among experts, and potential factors associated with expert judgments. Materials and methods Italian criminal cases were included when documentation from at least two expert evaluations was available. Data regarding the type of offense, the role of the expert authoring the report (public prosecutor’s consultant, defense consultant, court-appointed expert), the evaluation methodology, and the judgment of criminal responsibility or non-responsibility were collected in an Excel database. Descriptive statistical analyses were performed, followed by concordance analyses among evaluators regarding clinical and medico-legal diagnoses. All statistical analyses were conducted using IBM SPSS software. Results A total of 97 judicial cases were included, analyzed through 253 expert reports, 53 of which concerned homicide. In 72.2% of evaluations, only one expert was involved. Psychiatrists were the most frequently involved specialists (70.9%), followed by forensic physicians and psychologists. The clinical interview was supplemented by personality tests (39.5%), neuropsychological tests (28.3%), neuroimaging (23.5%), EEG (9.2%), and toxicological analyses (8.3%). The most frequent diagnoses were personality disorders, depressive disorders, and schizophrenia spectrum disorders. Agreement between the public prosecutor’s consultant and the defense consultant in distinguishing healthy from pathological conditions was very low (κ = 0.126; N = 39), whereas agreement between the public prosecutor’s consultant and the court-appointed expert was good (κ = 0.685; N = 42). Overall agreement regarding criminal responsibility was very low (Krippendorff’s α = 0.196; 95% CI 0.026–0.358; N = 91). Pairwise analyses showed no agreement between the public prosecutor’s and defense consultants (κ = 0.013; 20.5% observed agreement) and weak agreement between the public prosecutor’s consultant and the court-appointed expert (κ = 0.384), indicating overall low and heterogeneous agreement among expert evaluators. Discussion The findings highlight substantial heterogeneity in training backgrounds and methodological approaches, as well as a low level of agreement among experts in both clinical and medico-legal diagnoses. This study, the first based on an Italian judicial case series, underscores the need to develop shared guidelines and more standardized and certifiable training pathways in order to ensure more reliable forensic psychiatric evaluations and to support judicial decision-making with stronger scientific grounding.
Introduzione L’imputabilità e il vizio di mente rappresentano tematiche complesse e ampiamente dibattute nell’ambito psichiatrico-forense. La mancanza di standard scientifici condivisi, unitamente alla possibile presenza di bias cognitivi, può influenzare gli accertamenti consulenziali e peritali, con potenziali ricadute sugli esiti processuali. Il presente studio si propone di analizzare, in una casistica giudiziaria italiana, la metodologia adottata da consulenti tecnici e periti, il grado di concordanza tra gli esperti e i potenziali fattori associati al giudizio consulenziale o peritale. Materiali e metodi Sono stati inclusi nello studio casi giudiziari italiani in ambito penale per i quali fosse disponibile documentazione relativa ad almeno due valutazioni consulenziali e/o peritali. I dati riguardanti il tipo di reato, il ruolo dell’esperto autore dell’elaborato (consulente tecnico del pubblico ministero - PM, consulente tecnico della difesa, perito), la metodologia valutativa adottata e il giudizio di imputabilità o non imputabilità sono stati raccolti in un database in formato Excel. È stata effettuata un’analisi statistica descrittiva dei dati, seguita da un’analisi di concordanza tra i diversi valutatori in relazione alla diagnosi clinica e medico-legale. Tutte le analisi statistiche sono state eseguite utilizzando il software IBM SPSS. Risultati Sono stati inclusi 97 casi giudiziari analizzati attraverso 253 elaborati consulenziali/peritali, di cui 53 relativi a omicidio. Nel 72,2% delle valutazioni era coinvolto un solo esperto; gli psichiatri hanno rappresentato la figura più frequente (70,9%), seguiti da specialisti in medicina legale e psicologi. Il colloquio clinico è stato integrato da test di personalità (39,5%), test neuropsicologici (28,3%), neuroimaging (23,5%), EEG (9,2%) ed esami tossicologici (8,3%). Le diagnosi più frequenti sono risultate disturbi di personalità, disturbi depressivi e disturbi dello spettro schizofrenico. La concordanza tra consulente del PM e della difesa nella distinzione sano/patologico è risultata molto bassa (κ = 0,126; N = 39), mentre tra PM e perito del giudice è risultata buona (κ = 0,685; N = 42). La concordanza complessiva sulla capacità di intendere e di volere è risultata molto bassa (α di Krippendorff = 0,196; IC95% 0,026–0,358; N = 91). L’analisi a coppie ha mostrato accordo nullo tra PM e difesa (κ = 0,013; 20,5%), e debole tra PM e perito (κ = 0,384), evidenziando una concordanza complessivamente bassa ed eterogenea tra le diverse figure peritali. Discussione I risultati evidenziano una marcata eterogeneità nei percorsi formativi e negli approcci metodologici, nonché un basso livello di concordanza tra periti e consulenti in ambito diagnostico clinico e medico-legale. Lo studio, primo su una casistica giudiziaria italiana, evidenzia la necessità di proporre e sviluppare linee guida condivise e di delineare percorsi formativi più uniformi e certificabili, al fine di garantire valutazioni psichiatrico-forensi più affidabili e di supportare le decisioni processuali con maggiore solidità scientifica.
Capacità di intendere e di volere in ambito penalistico: analisi della concordanza valutativa in psichiatria forense
DEBERTOLIS, CAMILLA
2025/2026
Abstract
Introduction Criminal responsibility and the insanity defense represent complex and widely debated issues in forensic psychiatry. The lack of shared scientific standards, together with the potential influence of cognitive biases, may affect expert evaluations, with possible implications for judicial outcomes. The present study aimed to analyze, within an Italian judicial case series, the methodology adopted by expert witnesses and consultants, the level of agreement among experts, and potential factors associated with expert judgments. Materials and methods Italian criminal cases were included when documentation from at least two expert evaluations was available. Data regarding the type of offense, the role of the expert authoring the report (public prosecutor’s consultant, defense consultant, court-appointed expert), the evaluation methodology, and the judgment of criminal responsibility or non-responsibility were collected in an Excel database. Descriptive statistical analyses were performed, followed by concordance analyses among evaluators regarding clinical and medico-legal diagnoses. All statistical analyses were conducted using IBM SPSS software. Results A total of 97 judicial cases were included, analyzed through 253 expert reports, 53 of which concerned homicide. In 72.2% of evaluations, only one expert was involved. Psychiatrists were the most frequently involved specialists (70.9%), followed by forensic physicians and psychologists. The clinical interview was supplemented by personality tests (39.5%), neuropsychological tests (28.3%), neuroimaging (23.5%), EEG (9.2%), and toxicological analyses (8.3%). The most frequent diagnoses were personality disorders, depressive disorders, and schizophrenia spectrum disorders. Agreement between the public prosecutor’s consultant and the defense consultant in distinguishing healthy from pathological conditions was very low (κ = 0.126; N = 39), whereas agreement between the public prosecutor’s consultant and the court-appointed expert was good (κ = 0.685; N = 42). Overall agreement regarding criminal responsibility was very low (Krippendorff’s α = 0.196; 95% CI 0.026–0.358; N = 91). Pairwise analyses showed no agreement between the public prosecutor’s and defense consultants (κ = 0.013; 20.5% observed agreement) and weak agreement between the public prosecutor’s consultant and the court-appointed expert (κ = 0.384), indicating overall low and heterogeneous agreement among expert evaluators. Discussion The findings highlight substantial heterogeneity in training backgrounds and methodological approaches, as well as a low level of agreement among experts in both clinical and medico-legal diagnoses. This study, the first based on an Italian judicial case series, underscores the need to develop shared guidelines and more standardized and certifiable training pathways in order to ensure more reliable forensic psychiatric evaluations and to support judicial decision-making with stronger scientific grounding.| File | Dimensione | Formato | |
|---|---|---|---|
|
Tesi_Camilla_Debertolis versione definitiva-.pdf
Accesso riservato
Dimensione
533.52 kB
Formato
Adobe PDF
|
533.52 kB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/104905