Self-Medication Following AI Advice among Adult Chatbot Users: A Cross-Sectional Study
DOI:
https://doi.org/10.37018/LLIF5841Keywords:
Self Medication, Artificial Intelligence, Chatbots, Digital Health, Information Seeking BehaviorAbstract
Background: AI chatbots are increasingly used for health information, but the accuracy and safety of their recommendations may vary. Evidence is limited on whether adults act on chatbot-generated health advice in Pakistan. The objective of this study was to determine the prevalence of following AI chatbot advice, compare trust in chatbot information across age groups, and identify associated attitudinal factors.
Methodology: This cross-sectional study was conducted by the Department of Medicine, Allama Iqbal Medical College, Lahore, from September 2025 to March 2026. Adults aged 18 years or older residing in Pakistan, with internet access, English proficiency, and previous AI chatbot use, were recruited through social media using a structured online questionnaire. Data from 702 participants were analysed using R version 4.5.1. Descriptive statistics, chi-square tests, Kruskal-Wallis tests, Spearman correlation, and multivariable logistic regression were used. Ethical approval and electronic informed consent were obtained.
Results: Of 702 participants, 640 (91.2%) were aged 18 to 25 years, and 658 (93.7%) had used AI chatbots for health queries. The main reasons for use were convenience and speed (457, 65.1%), privacy (114, 16.2%), limited healthcare access (76, 10.8%), and trust in AI capabilities (55, 7.8%). Symptom checking was the most common query type (53.0%). Overall, 213 participants (30.3%; 95% CI 27.1–33.8) reported following a treatment or medication plan based solely on AI chatbot advice. Confidence in self-medication (OR 1.87, 95% CI 1.55–2.25; p<0.001) and trust in chatbot information (OR 1.31, 95% CI 1.05–1.64; p=0.018) were independently associated with following advice. Healthcare professionals remained the preferred source of health advice across age groups. Trust in chatbot information differed across age groups (H=9.46, p=0.024). Outcome and side-effect questions were not restricted to participants who reported following advice and therefore were not interpreted as outcomes in that subgroup.
Conclusion: Approximately one-third of participants reported following a treatment or medication plan based solely on AI chatbot advice. Confidence in self-medication and trust in chatbot information were associated with this behaviour. Healthcare professionals remained the preferred source of health advice, supporting the need for responsible use of AI-generated health information.
References
1. Yun HS, Bickmore T. Online health information-seeking in the era of large language models: cross-sectional web-based survey study. J Med Internet Res. 2025;27. doi:10.2196/68560.
2. Semigran HL, Linder JA, Gidengil C, Mehrotra A. Evaluation of symptom checkers for self diagnosis and triage: audit study. BMJ. 2015;351. doi:10.1136/bmj.h3480.
3. Wallace W, Chan C, Chidambaram S, Hanna L, Iqbal FM, Acharya A, et al. The diagnostic and triage accuracy of digital and online symptom checker tools: a systematic review. NPJ Digit Med. 2022;5:118. doi:10.1038/s41746-022-00667-w.
4. Grossman S, Zerilli T, Nathan JP. Appropriateness of ChatGPT as a resource for medication-related questions. Br J Clin Pharmacol. 2024;90(10):2691-2695. doi:10.1111/bcp.16212.
5. Arellano Carmona K, Chittamuru D, Kravitz RL, Ramondt S, Ramírez AS. Health information seeking from an intelligent web-based symptom checker: cross-sectional questionnaire study. J Med Internet Res. 2022;24(8). doi:10.2196/36322.
6. Nikitha BS, Roopa K, Kynshi SL, Chauhan RS, Girish BS, Srinivasan R. Artificial intelligence and internet influence on drug utilization: exploring self-medication trends in South Indian pharmacy students. Intell Pharm. 2024;2(6):814-820. doi:10.1016/j.ipha.2024.06.001.
7. Saifullah M, Sharafi S, Ghanei Gheshlagh R. Prevalence and patterns of self-medication in Pakistan: a systematic review and meta-analysis. BMC Public Health. 2026;26:232. doi:10.1186/s12889-025-25978-x.
8. Zafar SN, Syed R, Waqar S, Zubairi AJ, Vaqar T, Shaikh M, et al. Self-medication amongst university students of Karachi: prevalence, knowledge and attitudes. J Pak Med Assoc. 2008;58(4):214-217.
9. Mahdi S, Akram W, Mumtaz Z, Tariq A, Abdul-Majid, Akhtar A. AI chatbots and online health information use in self-medication among university students in Lahore, Pakistan: a cross-sectional study. J Health Wellness Community Res. 2026;4(8):1-10. doi:10.61919/txzsca54.
10. Goodman RS, Patrinely JR, Stone CA Jr, Zimmerman E, Donald RR, Chang SS, et al. Accuracy and reliability of chatbot responses to physician questions. JAMA Netw Open. 2023;6(10). doi:10.1001/jamanetworkopen.2023.36483.
11. Si Y, Yang Y, Wang X, Zu J, Chen X, Fan X, et al. Quality and accountability of ChatGPT in health care in low- and middle-income countries: simulated patient study. J Med Internet Res. 2024;26. doi:10.2196/56121.
12. Aboueid S, Meyer S, Wallace JR, Mahajan S, Chaurasia A. Young adults' perspectives on the use of symptom checkers for self-triage and self-diagnosis: qualitative study. JMIR Public Health Surveill. 2021;7(1). doi:10.2196/22637.
13. World Bank. Physicians (per 1,000 people) – Pakistan [Internet]. Washington (DC): World Bank; [cited 2026 Nov 4]. Available from: https://data.worldbank.org/indicator/SH.MED.PHYS.ZS?locations=PK.
14. Yang D, Nikoloski Z, Khalid G, Mossialos E. Pakistan's path to universal health coverage: national and regional insights. Int J Equity Health. 2024;23:162. doi:10.1186/s12939-024-02232-1.
15. World Health Organization. Ethics and governance of artificial intelligence for health: guidance on large multi-modal models. Geneva: World Health Organization; 2024. Available from: https://www.who.int/publications/i/item/9789240084759.
16. Ciecierski-Holmes T, Singh R, Axt M, Brenner S, Barteit S. Artificial intelligence for strengthening healthcare systems in low- and middle-income countries: a systematic scoping review. NPJ Digit Med. 2022;5(1):162. doi:10.1038/s41746-022-00700-y.
17. Wang X, Sanders HM, Liu Y, Seang K, Tran BX, Atanasov AG, et al. ChatGPT: promise and challenges for deployment in low- and middle-income countries. Lancet Reg Health West Pac. 2023;41:100905. doi:10.1016/j.lanwpc.2023.100905.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Muhammad Zubair, Muhammad Shahzaib, Muhammad Hamza Masood, Urvah Shafique, Muhammad Abdullah, Musa Masood, Ali Hassan, Muhammad Fazeel, Adeel Anjum, Shahzaib Rasool Randhawa, Murtijaz Abbas

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
The Journal of Fatima Jinnah Medical University follows the Attribution Creative Commons-Non commercial (CC BY-NC) license which allows the users to copy and redistribute the material in any medium or format, remix, transform and build upon the material. The users must give credit to the source and indicate, provide a link to the license, and indicate if changes were made. However, the CC By-NC license restricts the use of material for commercial purposes. For further details about the license please check the Creative Commons website. The editorial board of JFJMU strives hard for the authenticity and accuracy of the material published in the journal. However, findings and statements are views of the authors and do not necessarily represent views of the Editorial Board.