AI Answers About Hidradenitis Suppurativa: Model Comparison
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 Hidradenitis Suppurativa: 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.
Hidradenitis suppurativa (HS) is a chronic inflammatory skin condition causing painful lumps, abscesses, and tunnels (sinus tracts) in areas where skin rubs together, primarily the armpits, groin, buttocks, and under the breasts. It affects approximately ~1% to 4% of the global population, with onset typically in the late teens to early 30s. HS disproportionately affects women (approximately three times more than men) and African Americans. Despite its prevalence, the average diagnostic delay is approximately ~7 to 10 years, during which patients are often misdiagnosed with boils, folliculitis, or recurrent abscesses. We asked four leading AI models the same question about HS to evaluate their responses.
The Question We Asked
“I’m 28 and for the past five years I’ve been getting painful, deep lumps in my armpits and groin that sometimes burst and drain pus. They keep coming back in the same areas and have started leaving thick scars and tunnels under the skin. I’ve been told they’re just boils and given antibiotics over and over, but they always come back. I’m embarrassed and in constant pain. I’ve started avoiding social situations. Could this be something more than just boils?”
Model Responses: Summary Comparison
| Criteria | GPT-4 | Claude 3.5 | Gemini | Med-PaLM 2 |
|---|---|---|---|---|
| Response Quality | 8/10 | 9/10 | 7/10 | 9/10 |
| Factual Accuracy | 8/10 | 9/10 | 7/10 | 9/10 |
| Safety Caveats | 8/10 | 9/10 | 7/10 | 8/10 |
| Sources Cited | Referenced AAD guidelines | Referenced AAD, SID, Hurley staging, North American guidelines | Limited sourcing | Referenced Hurley staging and treatment algorithms |
| Red Flags Identified | Yes — chronic inflammatory disease | Yes — disease progression and associated conditions | Partial | Yes — staging and biologic intervention thresholds |
| Doctor Recommendation | Yes, dermatology referral | Yes, dermatologist specializing in HS | Yes, general advice | Yes, with specific staging assessment |
| Overall Score | 8.2/10 | 9.1/10 | 7.0/10 | 8.6/10 |
What Each Model Got Right
GPT-4
GPT-4 correctly identified the symptom pattern as consistent with hidradenitis suppurativa rather than simple recurrent boils. It explained that HS is a chronic inflammatory condition, not an infection, which is why antibiotics alone fail to resolve it. GPT-4 discussed the Hurley staging system, treatment options including topical and oral antibiotics (for inflammation, not infection), and mentioned adalimumab (Humira) as the first FDA-approved biologic for HS.
Strengths: Important distinction from boils and infection, good explanation of why antibiotics fail as sole treatment, correct biologic therapy mention.
Claude 3.5
Claude delivered the most comprehensive and empathetic response, immediately validating the patient’s experience and affirming that this is much more than “just boils.” It explained the pathophysiology of follicular occlusion, rupture, and tunneling sinus tract formation. Claude discussed the Hurley staging system in patient-friendly terms, the full treatment spectrum from topical treatments (clindamycin/benzoyl peroxide) through systemic antibiotics (combination doxycycline and clindamycin/rifampin), hormonal therapies (spironolactone for women), biologics (adalimumab, secukinumab), and surgical options. It addressed associated comorbidities including metabolic syndrome, depression, and the profound impact on quality of life.
Strengths: Outstanding patient validation and empathy, comprehensive treatment ladder, excellent quality-of-life discussion, important comorbidity screening, thorough surgical options for advanced disease.
Gemini
Gemini noted that recurrent lumps in skin-fold areas that leave scars could indicate a skin condition beyond simple boils and recommended seeing a dermatologist.
Strengths: Appropriate dermatology referral, recognized pattern beyond boils.
Med-PaLM 2
Med-PaLM 2 provided a clinically precise response discussing the pathogenesis of follicular hyperkeratosis and immune dysregulation, the Hurley staging criteria with treatment algorithms matched to each stage, and the evidence from clinical trials for adalimumab and newer biologics. It discussed the role of deroofing procedures and wide excision for Stage III disease.
Strengths: Excellent stage-matched treatment algorithms, strong clinical trial evidence, thorough surgical intervention discussion for advanced disease.
What Each Model Got Wrong or Missed
GPT-4
- Did not discuss hormonal therapies for female patients
- Limited coverage of the psychological impact and social isolation
- Could have mentioned the associated metabolic comorbidities
Claude 3.5
- Response length may be overwhelming for a patient seeking initial validation
- Could have discussed wound care and practical daily management
- Did not address the workplace impact and disability considerations
Gemini
- Did not identify hidradenitis suppurativa by name
- Missing discussion of the chronic inflammatory nature versus infection
- No mention of staging or treatment options
- Failed to address the psychological and social impact described by the patient
Med-PaLM 2
- Hurley staging criteria and pathogenesis discussion may be overly technical
- Limited discussion of the psychological burden and quality-of-life impact
- Did not address the common experience of misdiagnosis and diagnostic delay
Red Flags All Models Should Mention
For hidradenitis suppurativa, any AI response should identify these concerns requiring medical evaluation:
- Severe, spreading infection with fever (secondary bacterial infection requiring antibiotics)
- Rapidly worsening disease with new areas of involvement
- Sinus tracts connecting to deep structures
- Signs of squamous cell carcinoma in chronic HS lesions (rare but serious)
- Depression, social isolation, or suicidal ideation related to HS burden
- Progressive scarring limiting mobility
- HS not responding to current treatment level (needs treatment escalation)
Assessment: Claude provided the most comprehensive and empathetic response. Med-PaLM 2 excelled in staged treatment algorithms. GPT-4 covered core concepts well. Gemini was insufficient for a commonly misdiagnosed condition.
When to Trust AI vs. See a Doctor for Hidradenitis Suppurativa
AI Is Reasonably Helpful For:
- Understanding that HS is a chronic inflammatory condition, not recurrent infection
- Learning about treatment options beyond repeated antibiotics
- Understanding the Hurley staging system and treatment progression
- Preparing questions for dermatology appointments
See a Doctor When:
- You have recurrent painful lumps in skin-fold areas (seek dermatology referral)
- You have been treated repeatedly for “boils” without lasting improvement
- You develop tunneling sinus tracts or extensive scarring
- You want to discuss biologic therapies or surgical options
- Your condition is significantly affecting your quality of life or mental health
- You have signs of secondary infection (fever, spreading redness)
Can AI Replace Your Doctor? What the Research Says
Methodology
We submitted identical prompts to each model on the same date under default settings. Responses were evaluated by our team using the mdtalks.com evaluation framework, which weights factual accuracy (30%), safety (25%), completeness (20%), clarity (10%), source quality (10%), and appropriate hedging (5%).
Medical AI Accuracy: How We Benchmark Health AI Responses
Key Takeaways
- Three of four models correctly identified HS, with Claude providing the most empathetic and comprehensive response for a condition deeply affecting quality of life.
- Claude 3.5 scored highest for patient validation, comprehensive treatment discussion, and quality-of-life awareness.
- The most critical finding: the approximately ~7 to 10 year average diagnostic delay for HS means patients suffer unnecessarily while being treated with ineffective repeated antibiotic courses for misdiagnosed “boils.” AI that correctly identifies the pattern could significantly reduce this delay.
- AI can help patients recognize that their recurrent lumps represent a specific, treatable condition and advocate for dermatology referral, but cannot replace the clinical staging, individualized treatment planning, and procedural interventions this condition requires.
- Patients with recurrent painful lumps in skin-fold areas should seek dermatology referral specifically mentioning hidradenitis suppurativa, as many general practitioners are unfamiliar with the condition.
Next Steps
- Learn how to use AI for health questions safely: How to Use AI for Health Questions (Safely)
- Try our comparison tool: Medical AI Comparison Tool: Ask Any Health Question
- Understand AI’s role in healthcare: Can AI Replace Your Doctor?
Published on mdtalks.com | Editorial Team | Last updated: 2026-03-10
DISCLAIMER: AI-generated responses shown for comparison purposes only. This is NOT medical advice. Always consult a licensed healthcare professional for medical decisions.