Effect of a ChatGPT-based digital counseling intervention on anxiety and depression in patients with cancer: A prospective, randomized trial

IF 7.6 1区医学 Q1 ONCOLOGY

European Journal of Cancer Pub Date : 2025-04-03 DOI:10.1016/j.ejca.2025.115408

Orhun Akdogan , Galip Can Uyar , Enes Yesilbas , Kadriye Baskurt , Nuri Alperen Malkoc , Nuriye Ozdemir , Ozan Yazici , Berna Oksuzoglu , Aytug Uner , Ahmet Ozet , Osman Sutcuoglu

{"title":"Effect of a ChatGPT-based digital counseling intervention on anxiety and depression in patients with cancer: A prospective, randomized trial","authors":"Orhun Akdogan , Galip Can Uyar , Enes Yesilbas , Kadriye Baskurt , Nuri Alperen Malkoc , Nuriye Ozdemir , Ozan Yazici , Berna Oksuzoglu , Aytug Uner , Ahmet Ozet , Osman Sutcuoglu","doi":"10.1016/j.ejca.2025.115408","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><div>Psychological distress is prevalent among newly diagnosed cancer patients, often exacerbating treatment-related anxiety and depression. Artificial intelligence (AI)-driven interventions, such as large language models (LLMs), offer scalable solutions for patient education. However, their effectiveness in mitigating psychological distress in oncology remains unverified.</div></div><div><h3>Methods</h3><div>We conducted a prospective, two-center, randomized controlled trial between July and October 2024 to evaluate the effect of a ChatGPT-based digital counseling intervention on anxiety and depression in chemotherapy-naïve cancer patients. Participants were randomized 1:1 to either a ChatGPT-assisted counseling group or a standard clinician-led education group. The primary outcome was the change in Hospital Anxiety and Depression Scale (HADS) scores before the second chemotherapy cycle. Secondary outcomes included patient engagement behavior and the accuracy of AI-generated responses. Multivariable logistic regression assessed independent predictors of anxiety and depression reduction. The trial is registered with ClinicalTrials.gov (NCT06854315).</div></div><div><h3>Findings</h3><div>Of 196 screened patients, 160 were randomized (ChatGPT: n = 80; control: n = 80). The final cohort comprised 150 patients (median age: 64 years; 53.3 % female). At follow-up, the ChatGPT group exhibited significantly lower median HADS-Anxiety (6 [0–17] vs. 8 [1–17]; P = 0.002) and HADS-Depression (6 [0–20] vs. 9 [0–19]; P < 0.001) scores compared to controls. ChatGPT use was the strongest predictor of anxiety reduction (OR, 18.684; 95 % CI, 3.981–87.685; P < 0.001). AI responses were deemed appropriate in 88.5 % of cases, with inaccuracies most common in dietary and sexual health recommendations.</div></div><div><h3>Interpretation</h3><div>A ChatGPT-based counseling intervention significantly reduced anxiety and depression among newly diagnosed cancer patients prior to chemotherapy. These findings support the potential integration of AI-driven digital tools as adjuncts to conventional oncology education, enhancing patient support and engagement. Further research is needed to refine AI implementation in clinical practice.</div></div>","PeriodicalId":11980,"journal":{"name":"European Journal of Cancer","volume":"221 ","pages":"Article 115408"},"PeriodicalIF":7.6000,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"European Journal of Cancer","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0959804925001893","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ONCOLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Background

Psychological distress is prevalent among newly diagnosed cancer patients, often exacerbating treatment-related anxiety and depression. Artificial intelligence (AI)-driven interventions, such as large language models (LLMs), offer scalable solutions for patient education. However, their effectiveness in mitigating psychological distress in oncology remains unverified.

Methods

We conducted a prospective, two-center, randomized controlled trial between July and October 2024 to evaluate the effect of a ChatGPT-based digital counseling intervention on anxiety and depression in chemotherapy-naïve cancer patients. Participants were randomized 1:1 to either a ChatGPT-assisted counseling group or a standard clinician-led education group. The primary outcome was the change in Hospital Anxiety and Depression Scale (HADS) scores before the second chemotherapy cycle. Secondary outcomes included patient engagement behavior and the accuracy of AI-generated responses. Multivariable logistic regression assessed independent predictors of anxiety and depression reduction. The trial is registered with ClinicalTrials.gov (NCT06854315).

Findings

Of 196 screened patients, 160 were randomized (ChatGPT: n = 80; control: n = 80). The final cohort comprised 150 patients (median age: 64 years; 53.3 % female). At follow-up, the ChatGPT group exhibited significantly lower median HADS-Anxiety (6 [0–17] vs. 8 [1–17]; P = 0.002) and HADS-Depression (6 [0–20] vs. 9 [0–19]; P < 0.001) scores compared to controls. ChatGPT use was the strongest predictor of anxiety reduction (OR, 18.684; 95 % CI, 3.981–87.685; P < 0.001). AI responses were deemed appropriate in 88.5 % of cases, with inaccuracies most common in dietary and sexual health recommendations.

Interpretation

A ChatGPT-based counseling intervention significantly reduced anxiety and depression among newly diagnosed cancer patients prior to chemotherapy. These findings support the potential integration of AI-driven digital tools as adjuncts to conventional oncology education, enhancing patient support and engagement. Further research is needed to refine AI implementation in clinical practice.

查看原文本刊更多论文

基于chatgpt的数字咨询干预对癌症患者焦虑和抑郁的影响：一项前瞻性随机试验

心理困扰在新诊断的癌症患者中很普遍，经常加剧与治疗相关的焦虑和抑郁。人工智能（AI）驱动的干预措施，如大型语言模型（llm），为患者教育提供了可扩展的解决方案。然而，它们在减轻肿瘤患者心理困扰方面的有效性尚未得到证实。方法我们于2024年7月至10月进行了一项前瞻性、双中心、随机对照试验，以评估基于chatgpt的数字咨询干预对chemotherapy-naïve癌症患者焦虑和抑郁的影响。参与者以1:1的比例随机分配到chatgpt辅助咨询组或标准临床医生主导的教育组。主要观察指标为第二次化疗周期前医院焦虑抑郁量表（HADS）评分的变化。次要结果包括患者参与行为和人工智能生成反应的准确性。多变量logistic回归评估焦虑和抑郁减轻的独立预测因子。该试验已在ClinicalTrials.gov注册（NCT06854315）。196例筛查患者中，160例随机分组(ChatGPT: n = 80；控制:n = 80)。最终队列包括150例患者(中位年龄：64岁；53.3 %女)。在随访中，ChatGPT组表现出明显较低的HADS-Anxiety中位数（6 [0-17]vs. 8 [1-17]）；P = 0.002)和HADS-Depression (6 [0-20] vs. 9 [0-19]；P <； 0.001)得分与对照组相比。使用ChatGPT是焦虑减少的最强预测因子(OR, 18.684；95 % ci, 3.981-87.685；P & lt; 0.001)。在88.5% %的病例中，人工智能反应被认为是适当的，而在饮食和性健康建议中最常见的是不准确。基于chatgpt的咨询干预可显著降低新诊断癌症患者化疗前的焦虑和抑郁。这些发现支持了人工智能驱动的数字工具作为传统肿瘤学教育的辅助手段的潜在整合，增强了患者的支持和参与。需要进一步研究以完善人工智能在临床实践中的应用。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

European Journal of Cancer 医学-肿瘤学

CiteScore

11.50

自引率

4.80%

发文量

953

审稿时长

23 days

期刊介绍： The European Journal of Cancer (EJC) serves as a comprehensive platform integrating preclinical, digital, translational, and clinical research across the spectrum of cancer. From epidemiology, carcinogenesis, and biology to groundbreaking innovations in cancer treatment and patient care, the journal covers a wide array of topics. We publish original research, reviews, previews, editorial comments, and correspondence, fostering dialogue and advancement in the fight against cancer. Join us in our mission to drive progress and improve outcomes in cancer research and patient care.