ChatGPT Tailored Vet Messaging?

Hello, Summarians!

This newsletter attempts to help us all cope with the deluge of information we face every day. It usually contains three summaries and citations related to animal health.

Here’s how I suggest using it: If you find any of the studies even remotely useful, throw them into your archive folder. If not, hit delete.

Please help us spread the word about VetSummary 🙂 

Genetic Basis for Disk Disease in Dogs

Hansen’s type I intervertebral disc disease (IVDD) is a prevalent and costly spinal disorder in dogs, particularly affecting chondrodystrophic breeds, with dachshunds experiencing the highest incidence. Approximately 18% to 25% of dachshunds develop symptomatic disc disease (SDD) over their lifetime, with veterinary treatment costs reaching millions to billions globally. Type I IVDD is associated with early disc degeneration, leading to intervertebral disc calcification and potential spinal cord compression. Research indicates that dachshunds with radiographically detectable disc calcifications (RDDC) are at a higher risk of developing SDD, and these calcifications have been shown to be heritable. The number of RDDC peaks between two and four years of age, which led to the development of a radiographic scoring system (K-n) to assess disc calcification severity. 

Data from multiple studies show a correlation between K-n scores and SDD risk, with higher K-n values linked to a significantly increased likelihood of disease. In response, Nordic countries and other regions implemented breeding schemes that discourage the use of high K-n dachshunds and encourage low K-n individuals to improve spinal health. Further genetic research identified a causal variant, a fibroblast growth factor-4 retrogene insertion (FGF4L2), which is dominant for both chondrodystrophy and Type I IVDD. Studies in Nova Scotia duck tolling retrievers suggested a gene dose effect, where heterozygotes for FGF4L2 had fewer calcified discs than homozygotes. In dachshunds, the FGF4L2 insertion has a high allele frequency (0.92), and while it has been incorporated into genetic testing panels, there has been little guidance on its role in breeding decisions. 

The study aimed to examine the relationship between FGF4L2 genotype and radiographic disc calcification scores in dachshunds to provide recommendations for breed clubs. The results showed that dachshunds with one or two normal (N) alleles had significantly lower K-n scores, with most falling into the K0–K2 range, associated with a substantially lower risk of SDD. Specifically, 75% of dachshunds with one N allele and 86% with two N alleles had a K0 score, indicating a much lower risk compared to the breed average. Almost no dachshunds with at least one N allele had a high K-n score (K5+), which is linked to a 69% risk of SDD and frequent euthanasia due to repeated disc herniation. The study also found that while dachshunds with two N alleles had slightly lower K-n scores than heterozygotes, this difference was not statistically significant due to the small sample size. 

These findings challenge the previous assumption that FGF4L2 acts purely dominantly for spinal degeneration. Instead, the data support a gene dose effect where having one or two normal alleles reduces the severity of disc calcification. This is particularly relevant for dachshund breeding programs, as the breed is nearly fixed for the FGF4L2 allele. Rather than avoiding FGF4L2-positive dogs altogether, a more feasible breeding goal would be to prioritize dachshunds with at least one N allele to gradually increase its frequency in the population. However, since only 22.5% of dachshunds in the study carried an N allele, maintaining genetic diversity is crucial, and selective breeding of FGF4L2 homozygotes with lower K-n scores may be necessary. 

The study population was likely skewed towards dogs selected for breeding, meaning they may have already undergone selection for lower K-n scores. Additionally, the presence of some low K-n dachshunds with two FGF4L2 copies suggests that other genetic or environmental factors influence disc calcification. Despite these limitations, the study confirms that dachshunds with one or no FGF4L2 copies have a significantly lower risk of SDD and suggests that incorporating both K-n and genotype into breeding decisions would be beneficial. Given the high frequency of FGF4L2 in dachshunds, aiming for FGF4L2/N offspring represents a practical short-term goal for improving spinal health in the breed. 

Sullivan S, Redden D, Hardeng F, Sundqvist M, Kutzler M. The relationship between radiographic disc calcification score and FGF4L2 genotype in dachshunds. J Vet Intern Med. 2025; 39(1):e17281. doi:10.1111/jvim.17281 

Bottom line — It may be useful to test breeding animals and potential individual pets

ChatGPT for Tailored Communication

Antimicrobial-resistant (AMR) infections pose a significant public health threat, projected to cause 10 million deaths annually by 2050. While AMR occurs naturally, human factors such as improper antimicrobial use in humans and animals, incomplete prescriptions, and inadequate regulations accelerate its rise. Behavior change through education and persuasion is a crucial strategy for mitigating AMR, but effective public health interventions must be tailored to different cultural and contextual factors. Utilizing behavior-change models like the Health Belief Model enhances the effectiveness of communication strategies, ensuring they resonate with diverse audiences. 

Culturally tailored public health campaigns improve engagement by addressing specific community beliefs and practices. The rapid growth of AI, particularly generative AI, has created new opportunities to enhance communication, making knowledge more accessible and adaptable to different literacy levels, languages, and geographic regions. AI has already been applied in veterinary medicine for tasks like diagnostic imaging and clinical documentation, but its role in creating AMR awareness content remains unexplored. This study aimed to assess the potential of generative AI, specifically ChatGPT, in developing AMR messaging across languages and audiences. 

The findings suggest that ChatGPT can bridge linguistic and cultural barriers in AMR communication, but inconsistencies across languages highlight the influence of training data quality. The effectiveness of AI-generated content depends on prompt quality, which varies based on the expertise of the user crafting them. The involvement of behavioral scientists and public health experts could improve the selection of appropriate models and enhance content accuracy. While previous studies have demonstrated AI’s ability to generate public health messages for low-literacy audiences and urology patients, this study extends the discussion by evaluating generative AI’s adaptability across multiple languages and communication formats. 

Key strengths of the study include the use of a standardized rubric for assessing content across different languages, providing insights into ChatGPT's versatility. Additionally, analyzing AI-generated content for various formats and media platforms enhances understanding of its potential in public health communication. Identifying language-specific limitations also informs future improvements in AI applications for AMR and One Health research. 

However, limitations of the study include a small sample size, subjective selection of AMR drivers and target audiences, and varying levels of expertise among evaluators, which may have influenced content assessment. Moreover, the behavior-change models were chosen by ChatGPT without input from behavioral scientists, potentially affecting messaging effectiveness. The absence of direct evaluation by target audiences further limits conclusions about the real-world impact of AI-generated messages. 

Ethical concerns, including misinformation risks, copyright issues, unreliable data sources, and legal implications, underscore the need for professional oversight in AI-generated public health messaging. Future research should expand evaluations to multiple reviewers per language, integrate demographic factors, and compare AI-selected behavior-change models with those chosen by experts. Increasing sample sizes, testing systematic prompt variations, and involving target audiences will enhance understanding of AI’s effectiveness in AMR communication. 

Overall, generative AI shows promise for developing AMR awareness content but requires careful oversight, strategic prompt design, and collaboration with experts to maximize impact. Ensuring ethical, accurate, and culturally appropriate AI-generated messaging is essential for responsible implementation in public health initiatives. 

Akinyede, O., Yustyniuk, V., Ochwo, S., Aworh, M., & Wilkins, M. (2025). Preliminary exploration of ChatGPT-4 shows the potential of generative artificial intelligence for culturally tailored, multilingual antimicrobial resistance awareness messaging. American Journal of Veterinary Research https://doi.org/10.2460/ajvr.24.09.0283 

Bottom line — Early results support this as a beneficial.

Why do Equine Vets Burnout?

The study highlights significant challenges in equine veterinary practice, emphasizing high rates of burnout and workforce attrition. Over half of recent equine veterinary graduates leave the field within five years, with a large portion of current practitioners nearing retirement age, raising concerns about future workforce shortages. Burnout, linked to long hours, physical demands, and workplace challenges, is particularly prevalent among equine veterinarians compared to their peers in other veterinary disciplines. 

Using the Conservation of Resources (COR) Theory as a framework, the study identifies key factors influencing job dissatisfaction and burnout. Condition resources, such as workplace circumstances, were the most frequently cited concerns. Many participants reported experiencing discrimination based on gender, appearance, age, or disability, underscoring the need for improved diversity, equity, inclusion, and belonging (DEIB) in the field. Predictable schedules, reduced work hours, and shared workloads were highlighted as potential solutions to improve job satisfaction. 

Energy resources, including pay and student loan debt, were significant stressors. Equine veterinarians typically earn less than those in other veterinary fields, contributing to job dissatisfaction and prompting some to switch to small animal practice. Despite rising starting salaries, the pay gap persists, exacerbating retention challenges. 

Personal resources, such as problem-solving skills and deriving satisfaction from patient care, positively influenced job satisfaction, while object resources like equipment were rarely cited as stressors, indicating sufficiency in this area. Female veterinarians, younger practitioners, and those with work-life balance challenges were identified as particularly vulnerable to burnout. 

Strengths of the study include its qualitative approach and diverse participant backgrounds, though limitations were noted in the underrepresentation of BIPOC and LGBTQ+ individuals. Future research could focus on these groups to understand their unique experiences. The study concludes that addressing DEIB issues, improving pay, incentivizing rural practice, and fostering work culture changes are essential to mitigate burnout and workforce attrition in equine veterinary medicine. These findings inform potential interventions to sustain the equine veterinary workforce and ensure the profession's future viability. 

Whitaker K, Burnette A, Tan J-Y, Graves M, Hunt J, Devine E, et al. Factors influencing equine veterinarians' job satisfaction and retention: A focus group study. Equine Vet J. 2025. https://doi.org/10.1111/evj.14467 

Bottom line — Great information.

Just putting things in perspective …