Introduction: Selective outcome reporting (SOR) is the bias in which, based on the nature and direction of the results obtained, only a portion of the outcomes registered and measured are then actually included in the study publication; SOR can lead to overestimation of a treatment's efficacy and underestimation of adverse effects, in trials and reviews that incorporate them. Psychedelic medicine, that branch of psychiatry that studies the application of psychedelic molecules as treatments for mental disorders, is a field of research that has grown exponentially, is close to a decisive breakthrough (several molecules are close to approval by the regulating agency) and is potentially very profitable. Objectives: to examine the prevalence and types of SOR in trials regarding three of the major psychedelic molecules (Esketamine, MDMA, and Psilocybin) that have already been approved or are close to approval; to examine, in addition, the association between SOR and pharmaceutical industry involvement in trials, and the publication rate of trials. Methods: An inception cohort of clinical trials, registered on Clinicaltrials.gov by March 2023, involving psychedelic molecules was used. From this we selected randomized controlled trials, which evaluated the efficacy of Esketamine, MDMA or Psilocybin (alone or in combination with another intervention), on participants with a diagnosis or symptoms of a mental disorder; trials that were still ongoing (as of July 2024) or never started have been excluded. Publications of these RCTs were then identified. Outcomes declared on the Clinicaltrials.gov registry were compared with those reported in publications, and the percentage of trials having at least one discrepancy regarding primary outcomes was calculated. RCTs were classified as having direct or indirect industry involvement, and the prevalence of SOR in the two subgroups was compared. Results: 58 RCTs were included in the selected sample. 13 of these 58 (22.4%) did not publish results in any form; another 6.9% only as press releases or conference posters, but not in indexed journals. Of 29 trials, peer-reviewed publications were identified. 2 were not evaluable because they were retrospectively registered or had undefined primary outcomes. In 8 of 27 (29.6%) evaluable trials, there was at least one discrepancy regarding the primary outcomes declared on Clinicaltrials.gov and those reported in the publication: in 2 out of 8 articles there was a change in the method of outcome assessment, and in all 8 articles there was a change in the timing of assessment. Only 3 out of 27 trials were found to have no direct industry involvement; the extremely small size of this subgroup resulted in the prevalence estimate of SOR being too unstable to allow robust and reliable statistical analysis. Conclusions: SOR is a phenomenon present in psychedelic medicine, affecting estimates of treatment efficacy and potentially the patients' health. Moreover, studies in which industry is not directly involved are almost absent in this area of research, a potentially problematic aspect. Various mandatory measures could and should be implemented by legislators and journal publishers to reduce the SOR bias.
Introduzione: la pubblicazione selettiva dei risultati (SOR) è il bias nel quale, in base alla natura e la direzione dei risultati ottenuti, solo una parte degli esiti registrati e misurati viene poi effettivamente inclusa nella pubblicazione dello studio; il SOR può portare alla sovrastima dell’efficacia di un trattamento e a sottostimare gli effetti avversi, nelle sperimentazioni e nelle revisioni che le incorporano. La medicina psichedelica, quella branca della psichiatria che studia l’applicazione di molecole psichedeliche come trattamenti per disturbi mentali, è un ambito di ricerca cresciuto esponenzialmente, prossimo ad una svolta decisiva (diverse molecole sono vicine all’approvazione da parte dell’organo legislatore) e potenzialmente molto remunerativo. Obiettivi: esaminare la prevalenza e le tipologie di SOR nelle sperimentazioni riguardo tre delle principali molecole psichedeliche (Esketamina, MDMA e Psilocibina), già approvate o vicine all'approvazione; esaminare, inoltre, l’associazione tra il SOR e il coinvolgimento dell’industria farmaceutica nelle sperimentazioni, e il tasso di pubblicazione delle sperimentazioni. Metodi: è stata usata una coorte di partenza di sperimentazioni cliniche, registrate su Clinicaltrials.gov entro marzo 2023, riguardanti molecole psichedeliche. Da questa abbiamo selezionato sperimentazioni controllate randomizzate, che valutavano l’efficacia di Esketamina, MDMA o Psilocibina (da sole o in combinazione con un altro intervento), su partecipanti con una diagnosi o sintomi di un disturbo mentale; sono state escluse le sperimentazioni ancora in corso (a luglio 2024) o mai cominciate. Sono poi state identificate le pubblicazioni di questi RCT. Gli esiti dichiarati sul registro Clinicaltrials.gov sono stati comparati con quelli riportati nelle pubblicazioni, ed è stata calcolata la percentuale di sperimentazioni aventi almeno una discrepanza riguardo gli esiti primari. Gli RCT sono stati classificati come aventi coinvolgimento dell’industria diretto o indiretto, ed è stata confrontata la prevalenza di SOR nei due sottogruppi. Risultati: nel campione selezionato sono stati inclusi 58 RCT. 13 di questi 58 (22.4%) non hanno pubblicato i risultati in alcuna forma; un altro 6.9% solo come comunicati stampa o poster di conferenze, ma non in riviste indicizzate. Di 29 sperimentazioni sono state identificate le pubblicazioni peer-reviewed. 2 non erano valutabili in quanto registrate retrospettivamente o aventi esiti primari non definiti. In 8 delle 27 (29.6%) sperimentazioni valutabili, c’era almeno una discrepanza riguardo gli esiti primari dichiarati su Clinicaltrials.gov e quelli riportati nella pubblicazione: in 2 articoli su 8 è avvenuta una modifica del metodo di valutazione degli esiti, e in tutti gli 8 articoli è avvenuta una modifica delle tempistiche di valutazione. Solo 3 sperimentazioni su 27 sono risultate non avere un coinvolgimento diretto dell’industria; la dimensione estremamente ridotta di questo sottogruppo ha fatto sì che la stima della prevalenza di SOR fosse troppo instabile per permettere un'analisi statistica robusta e attendibile. Conclusioni: il SOR è un fenomeno presente nella medicina psichedelica, con ripercussioni sulle stime di efficacia dei trattamenti e potenzialmente sulla salute dei pazienti. In questo ambito di ricerca, inoltre, sono pressoché assenti studi in cui l’industria non è direttamente coinvolta, aspetto potenzialmente problematico. Vari provvedimenti vincolanti potrebbero e dovrebbero essere adottati dai legislatori e dagli editori delle riviste per ridurre il bias di SOR.
Bias di pubblicazione selettiva dei risultati e coinvolgimento dell'industria farmaceutica in RCTs sull'efficacia di psichedelici per disturbi mentali. Una revisione sistematica
BONESSIO, ANDREA
2023/2024
Abstract
Introduction: Selective outcome reporting (SOR) is the bias in which, based on the nature and direction of the results obtained, only a portion of the outcomes registered and measured are then actually included in the study publication; SOR can lead to overestimation of a treatment's efficacy and underestimation of adverse effects, in trials and reviews that incorporate them. Psychedelic medicine, that branch of psychiatry that studies the application of psychedelic molecules as treatments for mental disorders, is a field of research that has grown exponentially, is close to a decisive breakthrough (several molecules are close to approval by the regulating agency) and is potentially very profitable. Objectives: to examine the prevalence and types of SOR in trials regarding three of the major psychedelic molecules (Esketamine, MDMA, and Psilocybin) that have already been approved or are close to approval; to examine, in addition, the association between SOR and pharmaceutical industry involvement in trials, and the publication rate of trials. Methods: An inception cohort of clinical trials, registered on Clinicaltrials.gov by March 2023, involving psychedelic molecules was used. From this we selected randomized controlled trials, which evaluated the efficacy of Esketamine, MDMA or Psilocybin (alone or in combination with another intervention), on participants with a diagnosis or symptoms of a mental disorder; trials that were still ongoing (as of July 2024) or never started have been excluded. Publications of these RCTs were then identified. Outcomes declared on the Clinicaltrials.gov registry were compared with those reported in publications, and the percentage of trials having at least one discrepancy regarding primary outcomes was calculated. RCTs were classified as having direct or indirect industry involvement, and the prevalence of SOR in the two subgroups was compared. Results: 58 RCTs were included in the selected sample. 13 of these 58 (22.4%) did not publish results in any form; another 6.9% only as press releases or conference posters, but not in indexed journals. Of 29 trials, peer-reviewed publications were identified. 2 were not evaluable because they were retrospectively registered or had undefined primary outcomes. In 8 of 27 (29.6%) evaluable trials, there was at least one discrepancy regarding the primary outcomes declared on Clinicaltrials.gov and those reported in the publication: in 2 out of 8 articles there was a change in the method of outcome assessment, and in all 8 articles there was a change in the timing of assessment. Only 3 out of 27 trials were found to have no direct industry involvement; the extremely small size of this subgroup resulted in the prevalence estimate of SOR being too unstable to allow robust and reliable statistical analysis. Conclusions: SOR is a phenomenon present in psychedelic medicine, affecting estimates of treatment efficacy and potentially the patients' health. Moreover, studies in which industry is not directly involved are almost absent in this area of research, a potentially problematic aspect. Various mandatory measures could and should be implemented by legislators and journal publishers to reduce the SOR bias.File | Dimensione | Formato | |
---|---|---|---|
Bonessio_Andrea.pdf
accesso aperto
Dimensione
1.95 MB
Formato
Adobe PDF
|
1.95 MB | Adobe PDF | Visualizza/Apri |
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/20.500.12608/79337