August 26, 2026

|

50 min read

Boots in the Mud, Eyes on Clean Data: 25 Years of Livestock Field Studies

Six Lessons on Getting Clean Data Out of Livestock Field Studies

Insights from Jonalee Meyer-Watkins, MS, who has spent 25 years in animal health field studies: sponsor-side at Pfizer Animal Health, 15 years leading clinical monitoring at Midwest Veterinary Services, and project-based monitoring across food and companion animal work. Shared during Prelude’s Animal Health Insights webinar series.

Livestock studies do not happen in clinics. They happen at working feedlots, dairies, horse farms, and farrowing barns, on operations that were never designed around a protocol and never will be. The animals arrive at 1 AM. The barn has no cell reception. The people you are counting on for clean data already had a full day of work before your study showed up.

Jonalee Meyer-Watkins has spent 25 years in that gap between the protocol and the pen. She started sponsor-side at Pfizer Animal Health, working protocol development, study reports, and FDA CVM submissions, beginning with Excede and with the CIDR insert used in cattle synchronization and breeding programs. She then spent 15 years at Midwest Veterinary Services leading the clinical monitoring unit and the team responsible for the quality of data collected in the field. She also ranches in southwest Montana and markets cattle through Superior Livestock, which means she has stood on both sides of every site visit she has ever made.

In this Animal Health Insights session, Prelude CEO Tommy Jackson sat down with her for an hour on what it actually takes to bring regulated-quality data out of the field. Six lessons stood out.

1. Cattle Don’t Read the Protocol

Her framing of the difference between companion and livestock research deserves to be quoted in full: “They don’t read the protocol. They don’t read a calendar, and they really don’t care what time it is.”

A companion animal study mostly lives inside clinic hours. A livestock study runs on the animals’ clock. On a BRD study, calves can get sick all at once and take you deep into a randomization in a single day. Trucks arrive at 1 AM and someone has to receive those animals. Monitors wake up to a string of texts at 6 AM on a Tuesday.

The lesson for study teams: an enrollment and monitoring plan built on office hours is fiction. Staff the off-hours, budget for the surge days, and design the data capture to work at the speed the field actually moves.

2. The Study Has to Fit Inside the Operation’s Day

Asked how she gets producers to say yes, she started with a correction: “These guys aren’t necessarily looking for more work. Our job is to go in and put that study into their daily life.”

In practice, that means telling a site to block off six hours for a visit before they sign anything, and tying study timelines to compensation so nobody discovers the real cost mid-study. When staffing is the blocker, sponsors sometimes co-fund a site hire to make participation workable. It also means respecting the technicians doing the collection: acknowledgment and fair compensation, in her experience, are what turn a reluctant tech into a reliable one.

Do it that way and the numbers follow. Her ask-to-yes rate recruiting livestock sites runs about 75 percent, in an industry where companion animal sponsors pay for conference booths just to recruit sites.

3. Your Protocol and Your Forms Drive Your Data

One draft protocol she monitored asked crews to measure the circumference of a steer’s foot, in the chute, on untamed cattle. Her team went back to the sponsor: the odds of an animal holding still for that, while everyone keeps their fingers, are low. The sponsor switched to evaluating swelling and heat, and the study got cleaner data because of it.

Her rule is blunt: “Your protocol and your forms drive your data.” Have people who have actually run studies on working operations pressure-test the protocol before it locks. In her words, it is worth “the absolute tediousness and pain of going back and redoing the protocol a couple of times.”

4. Offline Data Capture Is the Spec, Not a Feature

Asked whether tablets make monitoring more efficient, she answered as someone who has worked dairy barns with bad reception and swine barns with none: “Offline data capture is key. Period. The end.”

The details matter as much as the headline. Forms need to be multi-animal, not one form per animal, so a chute crew is never waiting on a tablet. Forms have to load fast while animals are being worked. And when the tablet gets back to Wi-Fi, the data syncs to the cloud with the audit trail intact, capturing both when the data was entered and when it synced. That is what keeps offline collection contemporaneous in CVM terms.

Her one operational warning: make sure somebody plugs the tablets in at night.

5. Paper Costs More Than It Looks

Paper feels fast at the chute. Then GCP arrives. Keeping paper attributable, legible, and contemporaneous means dating every sheet uniformly and initialing as you go, what she described as “signing your name 20 times in one day.” An EDC audit trail handles the date, the time, and the attribution in the background. And any paper form used as backup still has to be transcribed into the system afterward, which doubles the work.

Paper also has failure modes that are hard to put in an SOP. She once watched a colt eat study data off a clipboard. Sows in a farrowing barn will do the same.

6. Read the Site Before You Sign the Site

Good sites look good beyond the protocol’s selection criteria: facilities in working condition, experienced animal handlers, and a veterinarian who stands behind the team. The red flags announce themselves if you walk the operation: corrals tumbling into one another, downed posts, pens that have not been scraped in six weeks, and a records system going every which direction.

Her test travels well beyond livestock: “If they can’t keep their own cards put together, they’re not going to keep your data put together.”

The Questions the Audience Wanted Answered

The audience kept the Q&A full for the entire hour. Five threads added depth beyond the main takeaways.

Do livestock sites underestimate the time commitment?

Often, yes, and it shows up later as partial protocol compliance. Her prevention happens before enrollment: work through the site’s contracted veterinarian, lay out per-visit hour estimates (block off six hours for an eligibility day), and tie timelines to compensation. On a recent equine study, a site veterinarian reported mid-study that an active case added about two hours to his day. Her answer: correct, it does. Now you know, so plan and price for it.

Can offline capture really be contemporaneous?

Yes. On the systems she has used, data goes in chute-side as the veterinarian examines the animal, with a technician entering findings system by system. The audit trail records both when data was entered and when it synced, so the record stays contemporaneous in CVM terms even when the barn has zero reception. Her aside during this answer got the biggest laugh of the hour: “this is not an advertisement for Prelude for offline data capture. It is.”

What about language barriers with tablets versus paper?

Interpreters at the farm-manager level, and simpler forms. Years ago her team proposed bilingual paper capture forms for a dairy study, and CVM declined them over dialect variation within Spanish. In practice, most Spanish-speaking crew members read English well, the lead farm manager is usually bilingual, and keeping forms simple carries the rest.

How do you keep a multi-year study consistent?

Write down every decision starting with the first animal: however you handle the first problem at the first site becomes the rule at every site. Keep a living FAQ, keep the project manager informed fast, and pre-schedule visits as far as a year out so sites can plan around you.

Small livestock are hard to count. How do you keep the data clean?

Piglets, in her words, “multiply without knowing.” Her reframe for frustrated crews has never failed her: “I know this does not make sense in real life. But at this moment, we’re not in real life, we’re in FDA land.” Tie every tedious reconciliation step back to what CVM requires, and crews will roll their eyes and then do it right, because the alternative is nullifying the case they just worked.

What This Means for Your Next Livestock Study

Every lesson in this hour points at the same order of operations. Site selection is data strategy, so walk the operation first. Protocol design is data strategy, so let operators pressure-test it before it locks. And capture is data strategy, so spec offline collection, multi-animal forms, and a background audit trail before the first animal enrolls, not after the first reconciliation crisis.

Her closing answer explained why any of it matters. Asked what she would put on a billboard for the animal health community, she answered as a rancher whose own calves ship to a feedlot this October: “We’re still feeding you.” The rancher matters. Animal health matters. Clean field data is how both get protected.


Watch the Full Conversation

This article is based on Prelude’s Animal Health Insights webinar featuring Jonalee Meyer-Watkins, a 25-year veteran of animal health field studies. Watch the full replay here.

Subscribe to our newsletter and stay up to date on the latest.

Sign up for our newsletter

Related resources