A new study has suggested that OpenAI’s AI chatbot, ChatGPT, may outperform primary care doctors in adhering to recognized treatment standards for clinical depression while avoiding the gender and social class biases sometimes present in doctor-patient relationships.
Published in the open-access journal Family Medicine and Community Health, the study reveals that ChatGPT can provide rapid, objective, data-driven insights that complement traditional diagnostic methods, all while maintaining confidentiality and anonymity.
The researchers assessed the AI technology’s ability to recommend appropriate treatment approaches for mild and severe major depression, examining whether these suggestions were influenced by biases related to gender or social class.
When asked, “What do you think a primary care physician should suggest in this situation?” ChatGPT’s responses included “watchful waiting,” “referral for psychotherapy,” “prescribed drugs (for depression/anxiety/sleep problems),” and “referral for psychotherapy plus prescribed drugs.”
The study found that just over 4% of family doctors recommended referring patients for psychotherapy in mild cases, as per clinical guidelines. In contrast, ChatGPT-3.5 and ChatGPT-4 suggested this option in 95% and 97.5% of cases, respectively.
Most physicians recommended either drug treatment exclusively (48%) or a combination of psychotherapy and prescribed drugs (32.5%). In severe cases, a majority of doctors recommended psychotherapy combined with prescribed drugs (44.5%).
ChatGPT proposed this treatment plan more frequently than doctors (72% for ChatGPT-3.5 and 100% for ChatGPT-4, in accordance with clinical guidelines). In contrast, 40% of doctors recommended drug treatment exclusively, an approach not endorsed by either ChatGPT version.
Additionally, doctors frequently recommended a combination of antidepressants, anti-anxiety medications, and sleeping pills (67.5% of cases). ChatGPT was more likely than doctors to recommend antidepressants alone, with 74% for version 3.5 and 68% for version 4.
ChatGPT-3.5 (26%) and ChatGPT-4 (32%) also suggested a combination of antidepressants, anti-anxiety drugs, and sleeping pills.
The researchers noted that “ChatGPT-4 demonstrated greater precision in aligning treatment with clinical guidelines. Furthermore, no discernible biases related to gender and socioeconomic status were detected in the ChatGPT systems.”
They concluded that ChatGPT has the potential to enhance decision-making in primary healthcare.
This study highlights the potential of AI chatbots like ChatGPT to assist healthcare professionals in providing evidence-based and unbiased treatment recommendations for clinical depression, ultimately improving patient care.
Source - IndiaTV
- Agencies
