Three separate stories moved through Washington simultaneously last week, each with implications for day-to-day veterinary practice.

On March 26, the Senate Judiciary Committee voted to advance the Combating Illicit Xylazine Act, bringing a three-year legislative effort closer to the finish line than it has ever been. The bill would classify xylazine as a Schedule III drug under the Controlled Substances Act while including specific statutory protections for veterinary access, which would amend the definition of "ultimate user" so that veterinarians, farmers, and ranchers remain clearly protected.

The stakes are straightforward. Xylazine is an essential sedative, especially for cattle. If the DEA were to schedule it through its own administrative process, those exemptions would not exist. The bill threads a narrow needle: law enforcement gets new tools to address the spread of the drug as an illicit substance, and the veterinary profession keeps access to a drug it cannot replace. Both the Biden and Trump administrations have supported it, the DEA and DOJ are on board, and on March 31, a bipartisan coalition of 41 state attorneys general sent a letter to House and Senate leadership urging immediate passage.

One dissenting scientific voice is worth knowing about. At a concurrent House Energy and Commerce hearing, a University of North Carolina pharmacologist testified that xylazine's presence in the illicit drug supply is already declining, being replaced by medetomidine, a different veterinary sedative, and that scheduling one drug may simply accelerate adoption of the other. Either way, the practical advice is the same: have a conversation with your distributors now, because xylazine access is likely to look different in the near future regardless of which path prevails.

USDA Secretary Brooke Rollins signaled last week that the U.S. is now evaluating a phased reopening of the Mexican border to feeder cattle imports, a meaningful shift from the blanket closure in place since the NWS outbreak began moving north. The Arizona ports are the first candidates, with active screwworm cases roughly 800 miles away. The Texas border, where cases have been detected within 200 miles, remains firmly closed.

As the AVMA covered earlier this year, the USDA is currently releasing 100 million sterile flies per week in Mexico, and a groundbreaking for the $1 billion domestic sterile fly production facility in South Texas is expected in mid-April. Once operational, it will produce up to 300 million sterile flies per week.

The concern for large animal practitioners is real and timely. NWS fly activity peaks in warm, humid conditions which are arriving quickly. Wound surveillance and prompt reporting protocols matter now more than ever for livestock clients in border states. The USDA updates its case map twice weekly at screwworm.gov.

The Farm, Food, and National Security Act of 2026 cleared the House Agriculture Committee in mid-March and is now heading to the full Congress. Two provisions stand out for practitioners:

The Farm Bill has been in negotiations for years and there is no guarantee this version passes intact. But having veterinary priorities formally embedded in a bill that has cleared committee is meaningful progress.

A recent Forbes Business Council piece by Chewy President Mita Malhotra makes a case that will feel both obvious and overdue: veterinary medicine wins when it prioritizes the relationship over the transaction. The starting point is the modern pet parent. According to 2025 research from the Human Animal Bond Research Institute and Chewy Health, 97% of pet parents consider their pets family members, yet nearly 82% report challenges in understanding aspects of their pet's health. That gap is exactly where the client relationship either becomes an asset or breaks down. When clients feel genuinely informed and supported, adherence improves, preventive care increases, and trust compounds over time.

The article's core argument is that relationship-centered care isn't a soft cultural aspiration, it's a strategic advantage. In a market shaped by rising costs and price sensitivity, practices built on genuine trust have differentiation that is hard to replicate. The author's practical suggestion is to identify one "relationship signature moment" (ex. a 24-hour post-procedure check-in, or a welcome station for pets and owners at arrival) and build it into every visit with enough consistency that it becomes part of the practice's identity. Crucially, the piece connects this to workforce sustainability: the same operating conditions that produce better client relationships (clinical autonomy, reasonable workloads, flexible scheduling) are also the conditions that reduce burnout and improve retention.

The subtext is that the practices most vulnerable to losing the relational dimension are precisely those under pressure to shorten appointments, maximize throughput, and standardize protocols. That description fits a lot of the corporate consolidation story. Independent practices have real economic constraints right now, but the relational model may be one of the clearest structural advantages they still hold over high-volume, lower-continuity competitors.

Last week brought one of the more important veterinary AI papers I’ve seen in a while and one I have been waiting a long time to see. Ma et al., published in JAVMA tested commercial radiograph AI tools on real-world general practice cases. The result was… not flattering. Looking at six commercial veterinary radiology AI platforms, performance was variable and overall low to moderate on canine abdominal radiographs sourced from general practice. The authors’ conclusion: none of the platforms appeared suitable for clinical use in their current form.

These tools are often marketed as if they can help fill the gap when a radiologist is not immediately available. In theory, that sounds reasonable. But it is something I have been arguing vehemently against for years. This paper suggests that they are not good enough to serve as screening tests and not reliable enough to function as the reassuring support many veterinarians hope they are getting. Across platforms, label-based sensitivity was low, meaning the models frequently missed the actual radiographic findings present in the case. And when the authors looked at a critical diagnosis like small intestinal obstruction, sensitivity ranged from just 23% to 69%. So if you are worried about an obstruction, send it to AI and get back a clean read… what then? Why even send it?

All of this comes back to transparency. These systems are trained on proprietary datasets. We usually do not know anything about where the data came from, how representative it is, how often the models are updated, or whether performance drifts over time. Even groups that claim they have transparency have confusing metrics which don’t align with this paper. That means clinics are being asked to trust black boxes without the level of disclosure that should be standard for any system influencing medical decisions. And frankly, it should not be the job of academics to keep independently stress-testing these product, and then telling the profession what the companies should have disclosed up front.

So where does that leave us? For now, exactly where many cautious veterinarians suspected: these tools are not ready for autonomous use in general practice, and they are not yet dependable enough to be treated as meaningful diagnostic support This is exactly what we suggested in the ACVR/ECVDI position statement on AI. If these tools want a place in veterinary medicine we need to have serious conversations about how they are used. Radiograph interpretations direct to GP is the fastest way to make money. But its not definitively the way to the best care. We need external validation, ongoing monitoring, transparent reporting, and proof that the tool helps on messy real-world cases.

-RBA

Trupanion has released a new study analyzing more than 71,000 pet insurance claims, and the trend line on toxicity is heading in the wrong direction. Since 2020, toxicity-related claims in the U.S. have increased by more than 30% in dogs and more than 50% in cats. The top culprits across both countries are grapes and raisins, chocolate, and medications -- both human pharmaceuticals and veterinary drugs. Xylitol, while less frequent, produces some of the most expensive cases, averaging more than $1,300 CAD per claim.

The numbers likely reflect a combination of factors: more pets in more households since 2020, growing insurance penetration pulling previously unreported cases into the dataset, and a genuinely expanded landscape of household hazards. GLP-1 medications, broader cannabis access, and a wider range of prescription pharmaceuticals in homes create exposure risks that simply weren't as common five years ago. Whatever the cause, the clinical picture is consistent with what most practitioners are already seeing. Client education on the core offenders remains one of the higher-yield prevention conversations in the exam room, and pointing clients toward the ASPCA Animal Poison Control Center (888-426-4435) as a standing resource is a small step that can meaningfully shorten response time when something goes wrong.

Here are some of the other stories that caught our eye and we're following this week from around the veterinary world and animal kingdom: