By Sagar Shankaran, Founder of CallSphere
AI voice agents call pet owners post-surgery to monitor recovery, catching complications 2.3 days earlier on average and reducing emergency readmissions by 34%.
Key takeaways
Every day, thousands of pets undergo surgical procedures at veterinary clinics across the country — spays, neuters, mass removals, orthopedic repairs, dental extractions, and exploratory surgeries. After the procedure, the standard discharge process involves handing the pet owner a sheet of post-operative instructions and saying "Call us if you have any concerns." Then the clinic moves on to the next patient.
This discharge-and-hope model has a fundamental flaw: pet owners are unreliable observers of post-surgical complications. Studies in veterinary surgery literature report that 8% to 12% of surgical patients experience complications, but pet owners often do not recognize early warning signs until complications have progressed to a more serious stage. A pet owner may not realize that mild redness around an incision site at day 2 is normal but increasing redness and swelling at day 5 indicates infection. They may not know that a brief period of reduced appetite after anesthesia is expected, but complete refusal to eat at 48 hours warrants a call.
The consequences of delayed complication detection are significant. A minor incision infection caught at day 3 requires a $50 antibiotic prescription. The same infection caught at day 7, after it has progressed to an abscess, requires a $400 to $800 re-sedation and surgical drain placement. An orthopedic implant loosening detected at the first week can be addressed with activity restriction; detected at week 3, it may require a $3,000 revision surgery.
Veterinary clinics know this gap exists. Many instruct their technicians to make follow-up calls at 24 and 72 hours post-surgery. But in practice, these calls rarely happen consistently. The same staffing pressures that affect the front desk affect the surgical team. Technicians are preparing for the next day's procedures, monitoring hospitalized patients, and assisting in consultations. Follow-up calls fall to the bottom of the priority list. Industry surveys suggest that fewer than 40% of veterinary practices consistently make post-surgical follow-up calls, and among those that do, fewer than 60% reach the pet owner on the first attempt.
Post-operative instruction sheets serve an important purpose, but they have well-documented limitations as a standalone safety net.
flowchart LR
CALLER(["Pet Parent"])
subgraph TEL["Telephony"]
SIP["Twilio SIP and PSTN"]
end
subgraph BRAIN["Veterinary AI Agent"]
STT["Streaming STT<br/>Deepgram or Whisper"]
NLU{"Intent and<br/>Entity Extraction"}
TOOLS["Tool Calls"]
TTS["Streaming TTS<br/>ElevenLabs or Rime"]
end
subgraph DATA["Live Data Plane"]
CRM[("CRM and Notes")]
CAL[("Calendar and<br/>Schedule")]
KB[("Knowledge Base<br/>and Policies")]
end
subgraph OUT["Outcomes"]
O1(["Visit booked"])
O2(["Refill called in"])
O3(["Emergency triage to staff"])
end
CALLER --> SIP --> STT --> NLU
NLU -->|Lookup| TOOLS
TOOLS <--> CRM
TOOLS <--> CAL
TOOLS <--> KB
NLU --> TTS --> SIP --> CALLER
NLU -->|Resolved| O1
NLU -->|Schedule| O2
NLU -->|Escalate| O3
style CALLER fill:#f1f5f9,stroke:#64748b,color:#0f172a
style NLU fill:#4f46e5,stroke:#4338ca,color:#fff
style O1 fill:#059669,stroke:#047857,color:#fff
style O2 fill:#0ea5e9,stroke:#0369a1,color:#fff
style O3 fill:#f59e0b,stroke:#d97706,color:#1f2937
Information overload at a stressful moment. Pet owners receive discharge instructions while simultaneously managing a groggy, disoriented animal in a noisy clinic environment. Retention of written medical instructions under stress is approximately 40% to 50% — a figure consistent across both human and veterinary medicine research.
Hear it before you finish reading
Talk to a live CallSphere AI voice agent for healthcare in your browser — 60 seconds, no signup.
Generic instructions miss breed-specific nuances. A standard post-spay instruction sheet cannot cover the different healing profiles of a 5-pound Chihuahua versus a 120-pound Great Dane. Brachycephalic breeds have different anesthesia recovery patterns. Certain breeds are predisposed to specific surgical complications.
No mechanism for proactive detection. Instructions tell the owner what to do if they notice a problem. They do not actively check whether a problem exists. A pet owner who is not looking for swelling will not find it until it becomes obvious — by which point the complication is more advanced.
The human tendency to minimize. Pet owners, particularly those who have been through surgery themselves, tend to normalize post-surgical symptoms. "She seems a little off, but that's normal after surgery, right?" This self-reassurance delays the call to the clinic by 24 to 48 hours on average.
CallSphere's post-surgical monitoring agent implements a structured follow-up protocol that makes proactive calls to pet owners at clinically significant intervals — typically 24 hours, 72 hours, and 7 days post-surgery. Each call follows a procedure-specific assessment script designed with veterinary surgical specialists.
Surgery Completed
│
▼
┌──────────────────────┐
│ Discharge + Instruct │
│ + AI Follow-Up Setup │
└──────────┬───────────┘
│
┌──────┼──────┬──────────────┐
▼ ▼ ▼ ▼
24 hr 72 hr 7 day As-Needed
Check Check Check (Triggered)
│ │ │ │
▼ ▼ ▼ ▼
┌────────────────────────────────────┐
│ Symptom Assessment Engine │
│ ┌─────────┐ ┌────────┐ ┌───────┐ │
│ │ Normal │ │ Watch │ │ Alert │ │
│ │ Recovery│ │ Closer │ │ Vet │ │
│ └─────────┘ └────────┘ └───────┘ │
└────────────────────────────────────┘
from callsphere import VoiceAgent, FollowUpScheduler
from callsphere.veterinary import SurgeryProtocol, RecoveryAssessment
# Define surgery-specific follow-up protocols
protocols = {
"spay_canine": SurgeryProtocol(
procedure="ovariohysterectomy",
species="canine",
checkpoints=[
{
"timing_hours": 24,
"questions": [
"Is your dog eating and drinking normally?",
"Has your dog vomited since coming home?",
"Is the incision site clean and dry?",
"Is your dog able to urinate and defecate?",
"Is your dog wearing the recovery cone?",
"On a scale of 1 to 10, how would you rate "
"your dog's energy level?"
],
"red_flags": [
"vomiting_persistent", "incision_open",
"bleeding_active", "not_urinating",
"extreme_lethargy", "pale_gums"
]
},
{
"timing_hours": 72,
"questions": [
"How is the incision site looking? Any redness, "
"swelling, or discharge?",
"Is your dog's appetite back to normal?",
"Is your dog trying to lick or chew at the "
"incision site?",
"Has your dog had normal bowel movements?",
"Is your dog more active than yesterday?"
],
"red_flags": [
"incision_swelling", "discharge_colored",
"fever_suspected", "appetite_absent",
"lethargy_worsening"
]
},
{
"timing_hours": 168, # 7 days
"questions": [
"Is the incision site healing well? Can you "
"describe what it looks like?",
"Is your dog fully back to normal energy "
"and appetite?",
"Have you been restricting activity as "
"instructed?",
"Do you have any concerns before the suture "
"removal appointment?"
],
"red_flags": [
"incision_not_healing", "sutures_missing",
"swelling_new", "behavior_change"
]
}
]
),
"dental_extraction": SurgeryProtocol(
procedure="dental_extraction",
species="canine",
checkpoints=[
{
"timing_hours": 24,
"questions": [
"Is your pet eating soft food?",
"Have you noticed any bleeding from the mouth?",
"Is your pet drooling excessively?",
"Is your pet able to drink water?"
],
"red_flags": [
"bleeding_ongoing", "not_drinking",
"facial_swelling", "extreme_pain_signs"
]
}
]
)
}
# Configure the follow-up agent
followup_agent = VoiceAgent(
name="Post-Surgery Recovery Agent",
voice="dr_sarah", # calm, caring tone
language="en-US",
system_prompt="""You are a post-surgery follow-up assistant
for {practice_name}. You are calling to check on a pet
that recently had surgery.
Your approach:
1. Identify yourself and the purpose of the call
2. Ask each recovery question from the protocol
3. Listen carefully for red-flag symptoms
4. Assess overall recovery trajectory
5. Provide reassurance for normal recovery signs
6. Escalate immediately if any red flags detected
CRITICAL RULES:
- NEVER say "everything is fine" — you are not a vet
- Say "that sounds like normal recovery" for expected symptoms
- For ANY concerning symptom, recommend calling the clinic
- For severe symptoms, offer to transfer immediately
- Document every response for the veterinary team
- Be empathetic — owners worry about their pets""",
tools=[
"assess_recovery_status",
"escalate_to_veterinarian",
"schedule_recheck_appointment",
"send_home_care_update",
"log_recovery_notes",
"transfer_to_surgical_team"
]
)
# Schedule follow-up calls at discharge
async def setup_post_surgical_followup(surgery_record):
"""Configure follow-up calls based on procedure type."""
protocol = protocols.get(
f"{surgery_record.procedure_type}_{surgery_record.species}",
protocols.get("general_surgery")
)
scheduler = FollowUpScheduler(agent=followup_agent)
for checkpoint in protocol.checkpoints:
call_time = surgery_record.discharge_time + timedelta(
hours=checkpoint["timing_hours"]
)
await scheduler.schedule_call(
phone=surgery_record.owner.phone,
scheduled_time=call_time,
context={
"pet_name": surgery_record.patient.name,
"procedure": surgery_record.procedure_description,
"surgeon": surgery_record.veterinarian.name,
"discharge_date": surgery_record.discharge_time.date(),
"medications": surgery_record.discharge_medications,
"activity_restrictions": surgery_record.restrictions,
"checkpoint": checkpoint
},
retry_policy={
"max_attempts": 3,
"retry_interval_hours": 2,
"escalate_on_no_answer": checkpoint.get(
"timing_hours") == 24
}
)
# Handle recovery assessment outcomes
@followup_agent.on_call_complete
async def handle_recovery_check(call):
assessment = RecoveryAssessment(call.responses)
if assessment.severity == "critical":
await notify_surgeon_immediately(
surgeon=call.metadata["surgeon"],
pet=call.metadata["pet_name"],
findings=assessment.summary,
owner_phone=call.caller_phone
)
elif assessment.severity == "concerning":
await schedule_early_recheck(
patient_id=call.metadata["patient_id"],
reason=assessment.summary,
urgency="next_available"
)
await send_enhanced_care_instructions(
phone=call.caller_phone,
instructions=assessment.care_adjustments
)
else:
await log_normal_recovery(
patient_id=call.metadata["patient_id"],
checkpoint=call.metadata["checkpoint"],
notes=assessment.summary
)
| Metric | Before AI Follow-Up | After AI Follow-Up | Change |
|---|---|---|---|
| Follow-up calls completed | 38% | 96% | +153% |
| Avg. days to complication detection | 5.1 days | 2.8 days | -45% |
| Emergency readmissions (surgical) | 7.2% | 4.8% | -33% |
| Revision surgery rate | 3.1% | 1.9% | -39% |
| Post-surgical complaint calls | 14/month | 4/month | -71% |
| Client satisfaction (surgical) | 72% | 93% | +29% |
| Technician hours on follow-up/week | 8 hrs | 0.5 hrs | -94% |
| Monthly savings (reduced readmissions) | $0 | $6,200 | New |
Week 1: Protocol Development. Work with your surgical team to define follow-up protocols for each procedure type your clinic performs. CallSphere provides evidence-based templates for common procedures (spay/neuter, mass removal, dental extraction, orthopedic repair, abdominal exploratory). Your veterinarians customize the questions and red-flag thresholds.
Week 2: Integration and Testing. Connect the follow-up system to your practice management system's surgical log. When a surgery is completed and discharge is processed, the follow-up sequence is automatically initiated. Test with staff members role-playing as pet owners to verify question flow and escalation triggers.
Week 3: Pilot Launch. Begin with one procedure type — typically spay/neuter, as it is the highest volume. Monitor every AI follow-up call for the first two weeks. Compare the AI's recovery assessments against the veterinarian's notes at suture removal appointments.
Still reading? Stop comparing — try CallSphere live.
See the healthcare AI agent handle a real call — complete, industry-specific, and live in your browser. No signup.
Week 4: Full Rollout. Expand to all procedure types. Configure surgery-specific protocols for orthopedic cases (which may require 6 weeks of follow-up calls), oncology cases, and complex procedures. Set up the surgeon notification workflow for red-flag escalations.
A high-volume surgical practice in Portland, Oregon — performing approximately 60 surgeries per week — deployed CallSphere's post-surgical follow-up agent in February 2026. Over the first 8 weeks, the agent completed 910 follow-up calls across 320 surgical patients. The agent flagged 47 cases for early clinical review, of which 38 were confirmed by veterinarians to benefit from the earlier intervention. The practice estimated that at least 12 of those cases would have progressed to complications requiring more intensive (and expensive) treatment without the proactive follow-up. Client satisfaction scores for surgical services rose from 74% to 94%, with many owners specifically mentioning the follow-up calls as a differentiator from other clinics.
The system retries up to three times at configurable intervals (typically every 2 hours). If no contact is made for the 24-hour post-surgical check — the most critical follow-up — the system escalates to the clinic's surgical team for manual follow-up. For later checkpoints, repeated no-answers trigger an SMS with a callback number. CallSphere tracks which owners consistently answer calls and optimizes call timing accordingly.
The current voice-based system focuses on verbal symptom assessment, which captures the majority of complications. For incision site assessment, the agent asks detailed descriptive questions about color, swelling, discharge, and odor. CallSphere is developing an integrated photo assessment feature that allows owners to text a photo of the incision during or after the follow-up call, which an AI image classifier evaluates and appends to the recovery notes.
When a pet has multiple procedures in the same surgical session (e.g., spay plus dental extraction plus mass removal), the follow-up protocol is composited from each individual procedure's checkpoint questions. The agent asks about each surgical site and procedure-specific recovery markers, and any red flag from any procedure triggers escalation. The questions are organized logically rather than repeated per procedure.
No. The AI follow-up calls complement, rather than replace, the in-person suture removal or recheck appointment. The goal is to catch complications between discharge and the recheck visit. Many clinics find that the follow-up calls actually increase recheck appointment compliance because owners feel more engaged in the recovery process and are reminded about the upcoming visit.
After every follow-up call, the attending veterinarian and surgical team receive a structured recovery report that includes the owner's responses to each question, the AI's severity assessment, any red flags detected, and the recommended action (normal monitoring, early recheck, or immediate contact). The report is attached to the patient's medical record in the practice management system and is available in the CallSphere dashboard.
Written by
Sagar Shankaran· Founder, CallSphere
Sagar Shankaran is the founder of CallSphere, where he builds production AI voice and chat agents deployed across healthcare, hospitality, real estate, and home services. He writes about agentic AI, LLM engineering, and shipping voice agents that handle real calls in production.
See how AI voice agents work for your industry. Live demo available -- no signup required.
A how-to for Colombian education and tutoring SMBs to answer parents and students instantly, book trial classes 24/7 in Spanish and English, and grow enrollment with a CallSphere AI agent.
Tbilisi professional-services firms serving relocating founders and IT companies use CallSphere AI voice and chat agents to answer enquiries 24/7 in English, Georgian and Russian and book consultations.
A practical how-to for Palau eco-resorts and dive operators on capturing every high-value, multilingual enquiry with a CallSphere AI voice and chat agent, while honouring Palau’s marine-conservation commitments.
How salons, spas and wellness SMBs across the UAE, Saudi Arabia and Qatar use CallSphere AI voice and chat agents to capture every booking 24/7 in Arabic, English and expat languages, and cut no-shows.
How estate agents and property managers in Luxembourg City and across the Grand Duchy use CallSphere to capture multilingual viewing and enquiry calls 24/7, GDPR compliant.
A practical guide for Senegalese logistics, freight, legal and consulting SMBs in Dakar and Thiès to capture every enquiry 24/7 with a CallSphere AI voice and chat agent in French, Wolof and English.
© 2026 CallSphere LLC. All rights reserved.
Made within New York
Watch how CallSphere handles real customer calls, schedules appointments, and processes payments — live.
Try Live DemoBook a DemoCalculate Your ROI