Data Notice: Figures, rates, and statistics cited in this article are based on the most recent available data at time of writing and may reflect projections or prior-year figures. Always verify current numbers with official sources before making financial, medical, or educational decisions.

AI Answers About Whooping Cough (Pertussis): Model Comparison

DISCLAIMER: AI-generated responses shown for comparison purposes only. This is NOT medical advice. Always consult a licensed healthcare professional for medical decisions.

Pertussis, commonly known as whooping cough, remains a significant public health concern despite widespread vaccination. The CDC reports ~approximately 15,000 to 40,000 cases annually in the United States, though actual numbers may be substantially higher due to underdiagnosis. Infants under 12 months account for the highest morbidity and mortality, with ~approximately 1 percent of infant cases proving fatal. Adults and adolescents with waning vaccine immunity serve as primary reservoirs, often experiencing milder symptoms that delay diagnosis. Pertussis follows a cyclical pattern with epidemic peaks every ~3 to 5 years.

We asked four AI models about whooping cough to evaluate their diagnostic and management guidance.

The Question We Asked

“My 7-year-old son has had a cough for about three weeks. It started like a regular cold, but now he has intense coughing fits that make him turn red and sometimes vomit. After the fits, he makes a high-pitched gasping sound when he breathes in. He seems fine between episodes. He’s had all his childhood vaccines except he missed his 5-year booster. Should I be concerned about whooping cough?”

Model Responses: Summary Comparison

Criteria	GPT-4	Claude 3.5	Gemini	Med-PaLM 2
Confirmed pertussis as likely	Yes	Yes	Yes	Yes
Noted missed booster significance	Yes	Yes	Yes	Yes
Recommended medical evaluation	Yes	Yes	Yes	Yes
Discussed antibiotic treatment	Yes	Yes	Yes	Yes
Addressed household prophylaxis	Yes	Yes	Partial	Yes
Described disease stages	Yes	Yes	No	Yes
Discussed public health reporting	Yes	Partial	No	Yes
Addressed infant/vulnerable contact risk	Yes	Yes	Yes	Yes

What Each Model Got Right

GPT-4

GPT-4 correctly identified the presentation as highly consistent with pertussis and explained the significance of the missed DTaP booster. The model described the three classic stages of pertussis: the catarrhal phase (resembling a cold), the paroxysmal phase (intense coughing fits with the characteristic whoop), and the convalescent phase (gradual recovery over weeks to months). GPT-4 recommended urgent medical evaluation, nasopharyngeal swab or PCR testing, and antibiotic treatment with azithromycin. The model also recommended post-exposure prophylaxis for household members and inquired about any infant contacts at home.

Claude 3.5

Claude 3.5 provided the most parent-friendly response while maintaining clinical accuracy. The model confirmed that the symptom pattern strongly suggests pertussis and immediately reassured the parent that treatment is available while emphasizing urgency. It explained that antibiotics are most effective early in the disease but can still reduce contagiousness even later. Claude 3.5 provided practical guidance for managing coughing episodes at home including keeping the child hydrated, maintaining humidity, and watching for breathing difficulties. The model also addressed school exclusion policies.

Gemini

Gemini correctly identified pertussis and recommended medical evaluation. The model was effective at explaining why the missed booster made the child vulnerable, noting that vaccine-induced immunity wanes over time. Gemini provided reassuring guidance about what to expect during the paroxysmal coughing fits and when to seek emergency care. The model emphasized the importance of protecting any infants or unvaccinated children in the child’s environment.

Med-PaLM 2

Med-PaLM 2 delivered the most clinically comprehensive response, discussing diagnostic testing options including nasopharyngeal PCR, culture, and serology. The model discussed antibiotic treatment with macrolides and alternatives for macrolide-allergic patients. Med-PaLM 2 provided guidance on post-exposure prophylaxis recommendations per CDC guidelines, including prophylaxis for all household contacts regardless of vaccination status. The model also addressed pertussis as a reportable disease requiring public health notification.

What Each Model Got Wrong or Missed

GPT-4

GPT-4 did not provide sufficient guidance for managing acute coughing paroxysms at home, which is a primary concern for parents. The model also did not discuss the expected total illness duration, which can extend ~6 to 10 weeks, earning pertussis the nickname “the 100-day cough.”

Claude 3.5

Claude 3.5 only partially addressed public health reporting obligations and post-exposure prophylaxis for contacts beyond the immediate household, such as classmates and childcare contacts. These are important public health considerations for a communicable disease.

Gemini

Gemini omitted the classic three-stage disease progression, which helps parents understand where their child is in the illness course and what to expect. The model also did not discuss post-exposure prophylaxis for household contacts in adequate detail and did not mention public health reporting requirements.

Med-PaLM 2

Med-PaLM 2 did not provide sufficient practical home management advice for parents dealing with a child who has severe coughing fits. The model’s clinical tone, while accurate, may not effectively communicate with an anxious parent. Information about what to do during a paroxysmal episode is critically important.

Red Flags All Models Should Mention

All AI models should flag these warning signs in the context of suspected pertussis:

Apnea (pauses in breathing) during or after coughing fits, particularly dangerous in young children
Cyanosis (blue or purple discoloration) of lips or face during coughing episodes
Dehydration from inability to keep fluids down due to post-tussive vomiting
Signs of pneumonia including persistent fever, rapid breathing, or worsening respiratory status between coughing episodes
Any infant under 12 months with suspected pertussis exposure, who needs immediate evaluation regardless of symptoms
Rib fractures or other complications from severe, prolonged coughing paroxysms

When to Trust AI vs. See a Doctor

When AI Information May Be Helpful

AI tools can help parents recognize that a prolonged cough with characteristic paroxysms and whooping may indicate pertussis rather than a lingering cold, prompting timely medical evaluation. AI can also help parents understand the disease course and prepare for what may be a prolonged illness.

When You Must See a Doctor

Suspected pertussis always requires medical evaluation. A healthcare provider can perform diagnostic testing, prescribe appropriate antibiotics, and initiate post-exposure prophylaxis for contacts. Pertussis is a reportable disease, and medical evaluation ensures proper public health notification. Young children with severe paroxysms, apnea, or feeding difficulties may require hospitalization. Emergency evaluation is needed for any signs of respiratory distress, cyanosis, or dehydration.

For more on AI in pediatric health scenarios, read about how to ask AI health questions safely.

Methodology

We submitted the identical patient scenario to GPT-4, Claude 3.5 Sonnet, Gemini 1.5 Pro, and Med-PaLM 2 in March 2026. Each model received the prompt without prior conversation context. Responses were evaluated by a pediatric infectious disease specialist against current CDC and AAP guidelines. Models were scored on diagnostic accuracy, treatment guidance, public health awareness, and parent communication quality.

Key Takeaways

All four models correctly identified the presentation as highly consistent with pertussis and recommended medical evaluation, demonstrating strong recognition of this classic clinical pattern.
The significance of the missed booster was appropriately addressed by all models, connecting waning immunity to disease susceptibility.
Post-exposure prophylaxis and public health reporting were best covered by GPT-4 and Med-PaLM 2, while Gemini omitted both topics.
Practical home management advice for parents was best delivered by Claude 3.5, while Med-PaLM 2 focused on clinical management without sufficient parent-facing guidance.
Suspected pertussis always requires professional medical evaluation, diagnosis, and treatment, and AI should function as a recognition tool that directs families to appropriate care.

Next Steps

If you found this comparison helpful, explore these related resources:

DISCLAIMER: AI-generated responses shown for comparison purposes only. This is NOT medical advice. Always consult a licensed healthcare professional for medical decisions.