Medical Errors are the Third Leading Cause of Death in the U.S.

Marshall, J.

Author correspondence: healthequityimagined@gmail.com

Cite this article: Marshall, J. (2025). Medical Errors are the Third Leading Cause of Death in the U.S. Diverse Perspectives on Wellness, 3(3), 1-5.

Abstract

A slight improvement in communication-related indicators that underlie positive hospital culture is associated with a significant reduction in adverse events, even as the response rate to errors remain about the same over the years.

Keywords: medical errors, hospital culture, artificial intelligence, machine learning

Introduction

Medical errors have been recently ranked as the third leading cause of death in the United States. 400,000 hospitalized patients experience some sort of preventable harm each year, with an estimated 200,000 deaths attributable to preventable medical errors that trigger adverse health conditions, life-ending consequences and cost the healthcare system $20 billion per year (Rodziewicz et al., 2024). Sensory input, working memory and long-term memory affect cognitive load differently. Once the volume of incoming information exceeds the processing capacity of the working memory, fatigue and cognitive overload set in, and factors like gender and insurance type increase the potential for errors in the ICU and psychiatric care, respectively. ⭃

8.9% of surgeons on one survey who were concerned that they had made a major medical error were significantly more likely to experience a lower mental quality of life, emotional exhaustion, depersonalization and symptoms of depression (Shanafelt et al., 2010). With 25% of most hospitals’ budgets going towards their administration, administrators with the capacity to evaluate the extent to which fatigue and cognitive overload create more opportunity for error have begun exploring alternatives to quality of care measures such as the 30 day re-admissions indicators used to distinguish between successful care and being readmitted at a later date due to continued health issues (Asgari et al., 2024). In the mid-2010’s, hospitals’ reliance on external analysts to process quality of care data and report them back to administrators often meant a 1–2 year data lag. Fast forward to the mid-2020’s, and AI’s capacity for performing more timely data reporting and procedural risk assessments while proposing diagnoses and treatment options has enabled real-time tracking based on quality of care indicators. The degree to which automated dictation and AI note-taking has mitigated some of the burnout from documentation burden means that greater initiatives focused on electronic health record input process improvement are on the horizon. ⭃

Decision-making algorithms

The advancement of computational powers, high-speed analysis, access to clean health care data and technology to create exact replicas of a person’s heart, liver, kidney and other specific organs down to the cellular level has fueled a shift towards health technology development (Faiyazuddin, 2025; Johnson, 2021; Palaniappan, Lin & Vogel, 2024). Administrative workflows where clinical notes based on the doctor’s and patient’s intuition are processed via machine-learning to inform diagnoses and predict mortality have demonstrated a high degree of accuracy when compared to data-driven decisions alone, and thus, the Food and Drug Administration has increasingly authorized machine learning programs for health care services (Chen, Szolovits & Ghassemi, 2019; Tulli, 2023). AI algorithms are showing better performance in detecting pneumonia in chest X-rays and malignant and nonmalignant skin lesions in dermatoscopic images when compared to radiologists, and their impact on care quality is largely dependent on a continued complementary co-existence with physicians (Chen, Szolovits & Ghassemi, 2019; Danks & LaRosa, 2018; Palaniappan, Lin & Vogel, 2024). ⭃

Methodology

The Agency for Healthcare Research and Quality (AHRQ) is a federal organization under the U.S. Department of Health and Human Services that develops health performance measurements to help hospitals measure quality of care. Between 2014 and 2017, AHRQ championed a reduction in hospital-acquired infections and adverse events, preventing 20,700 deaths and saving $7.7 billion. AHRQ’s Quality and Safety Review System (QSRS) is designed to identify the occurrence of adverse events that threaten patient safety in the hospital, and its Surveys on Patient Safety Culture Survey (SOPS) Hospital 2.0 Database hosts hospital-level data that indicates how each hospital’s culture of safety supports its capacity for quality care delivery. Values from each dataset show the percentage of healthcare administrators, physicians, providers and staff who answered “yes” to the survey questions used to inform each indicator. ⭃

Results

Higher percentages are indicative of a more positive hospital culture and increased patient safety, and Table 1 shows that just a slight improvement in communication-related indicators that underlie positive hospital culture is associated with a significant reduction in adverse events, even as the response rate to errors remain about the same over the years.

Discussion

Hospital culture underlies every aspect of the provider-patient dynamic, What we can glean from the gradual increase in communication about errors with little to no change in the error response rate is that a culture of open communication about medical errors supported by attribution systems that function independently from a hospital’s competitive infrastructure is an effective path towards improving a reliable patient care culture. Healthcare is still, and perhaps will always be staffed by empaths who celebrate their successes in reviving patients while carrying some of the emotional burden regarding unsuccessful treatment attempts, and this is true among administrators, physicians and data management personnel. Keeping the provider-patient dynamic free of misdiagnosis, incorrect procedures and unforeseen readmissions may require that administrations prioritize evolving machine-learning algorithms with updating medical, behavioral and social indicators. Patient anxiety regarding the incorporation of machines and robotics into their care means that both federal and in-house AI regulation to mitigate data breaches, data privacy risks will likely remain agile initiatives, and hospital administrators may find themselves increasingly synchronizing their operations in order to retain patient support and trust (Johnson, 2021).

References

Faiyazuddin, M., Rahman, S., Anand, G., Siddiqui, R., Mehta, R., Khatib, M. & Sah, R. (2025). The impact of artificial intelligence on healthcare: a comprehensive review of advancements in diagnostics, treatment, and operational efficiency. Health Science Reports, 8(1), e70312.

Johnson, K., Wei, W., Weeraratne, D., Frisse, M., Misulis, K., Rhee, K., Zhao, J. & Snowdon,, J. (2021). Precision Medicine, AI, and the Future of Personalized Health Care. Clinical Translational Science, 14(1), 86–93.

Patil, S., & Shankar, H. (2023). Transforming healthcare: harnessing the power of AI in the modern era. International Journal of Multidisciplinary Sciences and Arts, 2(2), 60–70.

Reddy, S., Fox, J., & Purohit, M. (2019). Artificial intelligence-enabled healthcare delivery. Journal of the Royal Society of Medicine, 112(1), 22–28.

Rodziewicz, T.L., Houseman, B., Vaqar, S., & Hipskind, J.E. (2024). Medical error reduction and prevention. StatPearls Publishing.

​Sahni, N., Stein, G., McKinsey, O., Zemmel, R., & Cutler, D. (2023). The Potential Impact of Artificial Intelligence on Healthcare Spending (Vol. 30857). Cambridge, MA, USA:: National Bureau of Economic Research.

Shanafelt, T. D., Balch, C. M., Bechamps, G., Russell, T., Dyrbye, L., Satele, D., … & Freischlag, J. (2010). Burnout and medical errors among American surgeons. Annals of surgery, 251(6), 995–1000.

Sunarti, S., Rahman, F., Naufal, M., Risky, M., Febriyanto, K. & Masnina, R. (2021). Artificial intelligence in healthcare: opportunities and risk for future. Gaceta sanitaria, 35, S67-S70.

Tulli, S. (2023). Analysis of the Effects of Artificial Intelligence (AI) Technology on the Healthcare Sector: A Critical Examination of Both Perspectives. International Journal of Social Trends, 1(1), 112–127.

Wu, Y., Wu, M., Wang, C., Lin, J., Liu, J., & Liu, S. (2024). Evaluating the prevalence of burnout among health care professionals related to electronic health record use: systematic review and meta-analysis. JMIR Medical Informatics, 12, e54811.

Yetgin, M. (2024). Perceptions of Participants Receiving Health Services About the Effects of AI in the Health Sectors. In Impact of AI and Robotics on the Medical Tourism Industry (69–98). IGI Global Scientific Publishing.

Acknowledgements

This research was supported by Health Analytics & Visualizations.

Conflict of Interest

The authors declare no conflict of interest or financial incentive. The authors’ relationships with the stakeholders and subject matter did not lead to unreasonable bias or compromise the objectivity of the research.

Rights and Permissions

Health Analytics & Visualizations supports the redistribution of its journal articles and content while also protecting its intellectual property from infringement. To request permission to republish our articles or content, email the Editor-in-Chief at info@stem4humanrights.com.