Opus-5.5 generated Doctor-Patient Conversations for All Human Diseases
8
2 commits
updated Sep 26, 2026
README
Opus-5.5 generated Doctor-Patient Conversations for All Human Diseases
The sequel to nisten/opus-doctor-patient-conversations-all-human-diseases (Opus 4.8). Same disease list, same 20-key schema, same ChatML conversations — regenerated from scratch with Claude Opus 5.5, one agent per disease, and held to a much stricter bar.
Covers every human disease listed on my previous work here: nisten/all-human-diseases
What changed since the Opus 4.8 set: the conversations are about 60% longer (avg 33 turns, ~5,700 tokens vs 29 turns, ~3,500) and run from presentation through workup, treatment and disposition with real vitals and labs, units and reference ranges included. Every PubMed ID cited in the text was checked live against NCBI before a record was accepted, and pubmed_refs is rebuilt per record from the citations the conversation actually makes (the curated source lists turned out to be mostly wrong: only 5.8% of their PMIDs point at the paper their title claims). Internal consistency was enforced, not just format: blood gases must satisfy Henderson-Hasselbalch, the anion gap must match the chemistry panel, doses must match the stated weight, and drugs must come from the record's own related_drugs.
Each conversation simulates a realistic patient from a randomly chosen diverse ethnic background, and in each one there is exactly one obfuscation or outright lie from the patient which the doctor tactfully catches and corrects. The drug-drug and drug-food interactions were all grounded via Health Canada Open Data: https://health-products.canada.ca/api/documentation/dpd-documentation-en.html
This is NOT a distillation, the agent traces are NOT included, only the final verified records are here, meant to be used as a source of truth and medical safety when training or using medical AI models & apps <3
- Records: 2,194 clinical encounters covering 2,194 unique human diseases
- Format: one JSON object per line (JSONL), 20 keys per record, 75 MB
- Conversations: avg 33.4 turns (25-59) and ~22,800 chars (~5,700 tokens) each; ~12.5M conversation tokens total
- Citations: 6,377 verified PubMed references across the records (4,198 distinct PMIDs), 2-5 per record, every one resolving on NCBI with a matching title
- Language: English
Provenance
- The seed list of disease names (in canonical order) is derived from nisten/all-human-diseases.
- The structured clinical fields (
description,related_drugs,drug_interactions,food_interactions,icd10,prevalence,body_systems) are copied verbatim from the same curated source as the Opus 4.8 set; the model never invents them. Two source texts were corrected after the fact: the methimazole-warfarin interaction description in three thyroid records (it had the INR direction backwards) and one trial summary (EMBARK's primary endpoint was not met). pubmed_refsis rebuilt, not verbatim. The curated source lists carried 3,925 distinct PMIDs; checked one by one against NCBI, 226 point at the paper their title claims, 3,582 resolve to an unrelated paper and 117 do not exist. Rather than ship that, each record'spubmed_refsnow lists the PMIDs the record itself cites, as{pmid, title, year}with the title taken from NCBI. Thesummaryfield of the old shape is gone (it was the fabricated part).- The 7 creative fields (personas, scenario, conversation, common mistakes, differential, related diseases, executive summary) were generated and then verified agentically by Claude Opus 5.5 subagents, one disease at a time. Each record passed automated gates before being committed: all 20 keys present, strict role alternation, >=15 turns, >=8,000-char conversation, >=3 items in every list field, zero emoji, no chatbot filler openers, no leaked generation instructions, and a live NCBI title check on every PMID in the text. Demographics were rolled per record from weighted workforce and population tables so clinician and patient backgrounds differ realistically.
- Two caveats, stated once: of the 2,195 names on the seed list one is excluded because the model's safety classifier stopped every generation attempt on it and I did not route around that, and the last six lines of the file were regenerated on a retry through the harness's generic
opusalias, so they are Opus-family but not verifiably 5.5. - The agentic traces / reasoning are NOT included — only the final, validated records.
Intended use
Built to be a source-of-truth dataset for RAG systems (clinical Q&A, retrieval-augmented assistants) and for training / fine-tuning small models on grounded, structured-yet-conversational medical dialogue. It is not medical advice, you're responsible for your own use of it.
Schema (20 keys)
| Key | Type | Source | Notes |
|---|---|---|---|
_type | string | — | always disease_chatml |
name | string | seed | canonical disease name |
aliases | string[] | source | synonyms |
search | string[] | source | search terms |
description | string | verbatim | clinical description |
related_drugs | object[] | verbatim | {din, brand_name, ingredient, atc_code, role} |
drug_interactions | object[] | verbatim | {drug_a, drug_b, severity, description} |
food_interactions | object[] | verbatim | may be empty for some diseases |
pubmed_refs | object[] | rebuilt | {pmid, title, year} — the PMIDs this record cites, title from NCBI |
icd10 | string | verbatim | ICD-10 code |
prevalence | string | verbatim | epidemiology text |
body_systems | string[] | verbatim | e.g. ["dermatologic"] |
source_disease | string | seed | name as keyed |
clinician_persona | string | generated | named physician, specialty, background |
patient_scenario | string | generated | named patient, presentation |
conversation | object[] | generated | ChatML-ready {role, content}, roles: system/user/assistant |
common_mistakes | object[] | generated | clinical pitfalls |
differential_diagnosis | object[] | generated | ranked differentials |
related_diseases | object[] | generated | adjacent conditions |
executive_summary | string | generated | concise case summary |
Disease coverage (2,194 unique diseases)
All disease names in the dataset, alphabetical, read top-to-bottom then left-to-right (scroll horizontally):
22q11.2 deletion syndrome Blastocystosis Constipation in children Fibrosis, pulmonary Incontinence, fecal Myopia Pressure sore Stings, jellyfish
A fib Blepharitis Contact dermatitis Fifth disease Incontinence, nighttime Myxofibrosarcoma Pressure ulcer Stings, scorpion
Abdominal aortic aneurysm Blocked tear duct Conversion disorder Filoviruses Incontinence, urinary NAFLD Preterm labor Stomach cancer
Abnormally excessive sweating Blocked ureter Cooley's anemia Fistula, arteriovenous Indigestion Nail fungus Priapism Stomach flu
Abscess, Bartholin's Blood in urine (hematuria) COPD Flatfeet Infant jaundice Narcissistic personality disorder Prickly heat Stomach pain, nonulcer
Absence seizure BMS Corns and calluses Flaviviruses Infant reflux Narcolepsy Primary adrenal insufficiency Stomach polyps
Acanthosis nigricans Body dysmorphic disorder Coronary artery disease Floaters Infantile hemangioma Nasal and paranasal tumors Primary biliary cholangitis Stomatodynia
Achalasia Body lice Coronary microvascular disease Floor of the mouth cancer Infectious arthritis Nasal polyps Primary biliary cirrhosis Stool holding
Achilles tendinitis Body odor and sweating Coronavirus disease 2019 (COVID-19) Floppy valve syndrome Infectious diseases Nasopharyngeal carcinoma Primary ciliary dyskinesia Strawberry hemangioma
Achilles tendon rupture Boils and carbuncles Corticobasal degeneration (corticobasal syndrome)Flu Infertility Nearsightedness Primary immunodeficiency Strep throat
Acid reflux Bone cancer Costochondritis Flu, avian Infertility, female Neck pain Primary lateral sclerosis (PLS) Stress cardiomyopathy
Acid reflux, infant Bone infection Costosternal chondrodynia Flu, bird Infertility, male Nephroblastoma Primary ovarian insufficiency Stress fractures
ACL injury Bone metastasis Costosternal syndrome Flu, swine Inflamed gallbladder Nephrogenic systemic fibrosis Primary polycythemia Stress incontinence
Acne Bone spurs Cough headaches Fluid around the heart Inflamed pancreas Nephrotic syndrome Primary progressive aphasia Stretch marks
Acne inversa Borderline personality disorder Cough, chronic FND Inflamed pericardium Neuralgia, postherpetic Primary sclerosing cholangitis Stroke
Acoustic neuroma Botulism COVID-19 Focal segmental glomerulosclerosis (FSGS)Inflammatory bowel disease (IBD) Neuralgia, trigeminal Proctitis Stump pain
Acquired immunodeficiency syndrome Bowel incontinence COVID-19-associated multisystem inflammatory syndrome in childrenFolate deficiency anemia Inflammatory breast cancer Neurilemmoma Progeria Stuttering
Acromegaly Bowel obstruction Crabs Folliculitis Influenza (flu) Neurinomatosis Progressive supranuclear palsy Stye (sty)
Actinic keratosis BPH Cradle cap Food allergy Influenza, avian Neuroblastoma Prolactinoma Subarachnoid hemorrhage
Acute coronary syndrome Brachial plexus injury Cramp, muscle Food allergy, egg Influenza, H1N1 Neurocardiogenic syncope Prolapse, mitral valve Subconjunctival hemorrhage (broken blood vessel in eye)
Acute flaccid myelitis (AFM) Bradycardia Cramps, menstrual Food allergy, milk Influenza, swine flu Neurodermatitis Prolapsed bladder Subfertility
Acute granulocytic leukemia Bradycardia-tachycardia syndrome Cranial arteritis Food poisoning Ingrown hair Neuroendocrine carcinoma of the skin Prolapsed uterus Substance abuse disorder
Acute inflammatory demyelinating polyneuropathyBrain aneurysm Craniopharyngioma Food-borne illness Ingrown toenails Neuroendocrine tumors Prostate cancer Sudden cardiac arrest
Acute kidney injury Brain arteriovenous malformation Craniosynostosis Foot drop Inguinal hernia Neurofibroma Prostate gland enlargement Sudden infant death syndrome (SIDS)
Acute liver failure Brain AVM (arteriovenous malformation) Creutzfeldt-Jakob disease Foot fracture Inherited metabolic disorders Neurofibromatosis type 1 Prostatitis Suicide
Acute lymphoblastic leukemia Brain metastases Crib death Foramen ovale Insomnia Neurofibrosarcoma Pruritis ani Suicide and suicidal thoughts
Acute lymphocytic leukemia Brain tumor Crohn's disease Fracture, arm Insulin resistance syndrome Neuroma, acoustic Pruritus Sun allergy
Acute lymphoid leukemia Brain tumor, child Croup Fracture, growth plate Intermittent claudication Neuroma, Morton's Pseudobulbar affect Sun poisoning
Acute myeloblastic leukemia Breast cancer CRPS Fracture, hip Intermittent explosive disorder Neuroma, plantar Pseudocholinesterase deficiency Sunburn
Acute myelogenous leukemia Breast cancer, inflammatory Cryoglobulinemia Fracture, leg Interstitial cystitis Neuromyelitis optica Pseudogout Sunken chest
Acute myeloid leukemia Breast cancer, male Cryptorchidism Fracture, stress Interstitial lung disease Neuromyelitis optica spectrum disorder Pseudomembranous colitis Supranuclear palsy
Acute nonlymphocytic leukemia Breast cysts CSF leak (Cerebrospinal fluid leak) Fractured nose Intestinal ischemia Neuropathy, autonomic Pseudotumor cerebri (idiopathic intracranial hypertension)Supraventricular tachycardia
Acute radiation sickness Breast infection CTCL Fractured rib Intestinal lipodystrophy Neuropathy, diabetic Psoriasis Suspicious breast lumps
Acute radiation syndrome Breast pain Curvature of the penis Frontal lobe epilepsy Intestinal obstruction Neuropathy, hereditary motor and sensoryPsoriatic arthritis SVT
Acute renal failure Broken ankle Curvature of the spine Frontal lobe seizures Intracranial hematoma Neuropathy, peripheral PTSD Sweating and body odor
Acute respiratory distress syndrome Broken arm Cushing syndrome Frontotemporal dementia Intracranial venous malformations Nevus Pubic lice (crabs) Sweating, abnormally excessive
Acute sinusitis Broken blood vessel in eye Cutaneous B-cell lymphoma Frontotemporal lobar degeneration Intraductal carcinoma Nickel allergy Puffy eyes Swelling
Addiction, alcohol Broken collarbone Cutaneous T-cell lymphoma Frostbite Intrahepatic cholestasis of pregnancy Nicotine addiction Pulmonary atresia Swimmer's ear
Addiction, gambling Broken foot Cutting/self-injury Frozen shoulder Intussusception Nicotine dependence Pulmonary atresia with intact ventricular septumSwimmer's itch
Addiction, nicotine Broken hand CVA FSGS Invasive lobular carcinoma Niemann-Pick disease Pulmonary atresia with ventricular septal defectSwollen knee
Addison's disease Broken heart syndrome CVID Fuchs dystrophy Iritis Nightmare disorder Pulmonary edema Swollen lymph nodes
Adenitis, mesenteric Broken hip Cyclic vomiting syndrome Functional dyspepsia Iron deficiency anemia Nighttime incontinence Pulmonary embolism Syndrome X
Adenomyosis Broken leg Cyclothymia (cyclothymic disorder) Functional neurologic disorder/conversion disorderIron overload NMO Pulmonary fibrosis Synostosis
ADHD in children Broken nose Cyclothymic disorder Functional neurological symptom disorderIrritable bowel syndrome NMOSD Pulmonary hypertension Synovial sarcoma
ADHD, Adult Broken ribs Cyst, Bartholin's Fungal infection, nail Ischemia, intestinal Noctural enuresis Pulmonary valve disease Syphilis
Adhesive capsulitis Broken toe Cyst, ganglion Funnel chest Ischemic bone necrosis Noise-related hearing loss Pulmonary valve stenosis Syringomyelia
Adjustment disorders Broken wrist Cyst, kidney GAD Ischemic colitis Non-Hodgkin's lymphoma PVCs Systemic exertional intolerance disease
Adnexal tumors Bronchiolitis Cyst, ovarian Galactorrhea Ischemic priapism Nonalcoholic fatty liver disease PVD Systemic lupus erythematosus
Adolescent schizophrenia Bronchitis Cyst, pancreatic Gallbladder cancer Islet cell cancer Nonallergic rhinitis PVPS Systemic mastocytosis
Adrenal cancer Brucellosis Cyst, pilonidal Gallbladder inflamation Itchy skin (pruritus) Nonischemic priapism Pyelonephritis Systemic sclerosis
Adrenal mass Brugada syndrome Cyst, spermatic Gallstones ITP Nonmelanoma skin cancer Pyloric stenosis Tachycardia
Adrenoleukodystrophy Bruxism (teeth grinding) Cystic fibrosis Gambling, compulsive Jaw tumors and cysts Nontropical sprue Pyoderma gangrenosum Takayasu's arteritis
Adult attention-deficit/hyperactivity disorder (ADHD)Buerger disease Cystitis Ganglion cyst Jellyfish stings Nonulcer dyspepsia RA Takotsubo cardiomyopathy
Adult Still disease Bulimia nervosa Cystitis, interstitial Gangrene Jet lag disorder Noonan syndrome Rabies Tapeworm infection
AFM Bullous pemphigoid Cystocele Gas and gas pains Jock itch Norovirus infection Radiation enteritis TAPVR
Age spots (liver spots) Bundle branch block Cysts, epidermoid Gastric cancer Jumper's knee Nose fracture Radiation sickness Tay-Sachs disease
Age-related macular degeneration, dry Bunions Cysts, sebaceous Gastric emptying, delayed Juvenile idiopathic arthritis Nymphomania Ramsay Hunt syndrome TB
Age-related macular degeneration, wet Bunyaviruses Cytomegalovirus (CMV) infection Gastric emptying, rapid Juvenile rheumatoid arthritis OAB Rapid gastric emptying TBI
Agnogenic myeloid metaplasia Burn injury Cytomegalovirus infection Gastritis Juvenile schizophrenia Obesity Rapid heartbeat Tears, decreased production
Agoraphobia Burning mouth syndrome Dandruff Gastroenteritis, viral Kaposi sarcoma Obesity, childhood Rash, poison ivy Teen depression
AIDP Burning thigh pain Daytime sleepiness Gastroesophageal reflux disease (GERD) Kawasaki disease Obsessive-compulsive disorder (OCD) Raynaud's disease Teeth grinding
AIDS/HIV Burns DCIS Gastrointestinal bleeding Keloid scar Obstetric cholestasis Reactive arthritis Temporal arteritis
Airplane ear Bursitis DDH Gastrointestinal stromal tumor (GIST) Keratitis Obstructive sleep apnea Reactive attachment disorder Temporal lobe seizure
Albinism Bursitis of the knee De Quervain tenosynovitis Gastroparesis Keratoconjunctivitis sicca Obstructive sleep apnea in children Reading disability, specific Temporomandibular disorders
Alcohol addiction C. difficile colitis Decreased tear production Gaucher disease Keratoconus Occupational asthma Rebound headache Temporomandibular joint disorders
Alcohol intolerance C. difficile infection Deep vein thrombosis (DVT) GBM Keratosis pilaris OCD Rectal cancer TEN
Alcohol poisoning Calciphylaxis Deficient color vision Gender dysphoria Keratosis, actinic Ocular albinism Rectal inflammation Tendinitis
Alcohol use disorder Calluses and corns Degenerative arthritis Gender identity disorder Keratosis, seborrheic Ocular rosacea Rectal itching Tendinitis, Achilles
Alcohol-associated hepatitis Cancer Dehydration Generalized anxiety disorder Ketoacidosis, diabetic Oculocutaneous albinism Rectal prolapse Tendinitis, patellar
Alcoholic hepatitis Cancer, acute lymphocytic leukemia Delayed ejaculation Genital herpes Kidney cancer Oculosympathetic palsy Rectal ulcer Tennis elbow
Allergic granulomatosis Cancer, anal Delayed gastric emptying Genital warts Kidney cysts ODD Rectovaginal fistula Tenosynovitis, de Quervain's
Allergic granulomatosis and angiitis Cancer, basal cell Delayed sleep phase Geographic tongue Kidney disease, chronic Odontogenic tumors and cysts Recurrent breast cancer Tension headache
Allergic rhinitis Cancer, bladder Delayed sleep phase syndrome GERD Kidney failure, acute OHSS Recurrent oral ulcers Testicle, retractile
Allergies Cancer, bone Delayed sleep-wake phase sleep disorder Germ cell tumors Kidney failure, chronic Olfactory neuroblastoma Red eye Testicle, undescended
Allergy, dust mite Cancer, breast Delirium German measles Kidney infection Oligodendroglioma Reflux, bile Testicular cancer
Allergy, egg Cancer, carcinoid tumors Dementia Gestational diabetes Kidney stones Onychomycosis Reflux, vesicoureteral Testicular torsion
Allergy, food Cancer, cervical Dementia, frontotemporal Giant cell arteritis Kissing disease Oppositional defiant disorder (ODD) Regurgitation, aortic valve Testosterone deficiency
Allergy, latex Cancer, chronic lymphocytic leukemia Dementia, Lewy body Giant lymph node hyperplasia Klatskin tumor Optic neuritis Regurgitation, mitral valve Tetanus
Allergy, milk Cancer, chronic myelogenous leukemia Dementia, vascular Giardia infection (giardiasis) Kleptomania Oral cancer Reiter's syndrome Tetralogy of Fallot
Allergy, mold Cancer, colon Dengue fever Gilbert syndrome Klinefelter syndrome Oral candidiasis REM sleep behavior disorder TGA
Allergy, nickel Cancer, endometrial Dengue hemorrhagic fever Gingivitis Klippel-Trenaunay syndrome Oral lichen planus Renal artery stenosis Thalassemia
Allergy, peanut Cancer, esophageal Dependence, nicotine Glaucoma Knee bursitis Oral thrush Renal cancer Thoracic aortic aneurysm
Allergy, penicillin Cancer, eye melanoma Depersonalization-derealization disorderGlioblastoma Knee pain Orchitis Renal failure, acute Thoracic outlet syndrome
Allergy, pet Cancer, gallbladder Depression (major depressive disorder) Glioblastoma multiforme KP Orthostatic hypotension (postural hypotension)Renal lithiasis Three-day measles
Allergy, shellfish Cancer, gastric Depression, postpartum Glioma Kyphosis Osteoarthritis Residual limb pain Throat cancer
Allergy, wheat Cancer, hairy cell leukemia Depression, teen Glomerulonephritis Lactase deficiency Osteoarthritis, cervical Respiratory syncytial virus (RSV) Thromboangiitis obliterans
Alopecia Cancer, Hodgkin's disease Dermatitis Glossodynia Lactation mastitis Osteochondritis dissecans Restless legs syndrome Thrombocytopenia
Alpha-gal syndrome Cancer, Hurthle cell Dermatitis, atopic Glucocerebrosidase deficiency Lactose intolerance Osteogenic sarcoma Retinal detachment Thrombocytosis
ALS Cancer, inflammatory breast Dermatitis, cercarial Gluten-sensitive enteropathy Laryngitis Osteomyelitis Retinal diseases Thrombophlebitis
Alveolar osteitis Cancer, kidney Dermatitis, contact Goiter Lassa fever Osteonecrosis Retinoblastoma Thrush, oral
Alzheimer's disease Cancer, leukemia Dermatitis, scratch Golfer's elbow Lateral elbow tendinopathy Osteophytes Retinopathy, diabetic Thumb arthritis
Ambiguous genitalia Cancer, lip Dermatitis, seborrheic Gonorrhea Lateral epicondylitis Osteoporosis Retractile testicle Thunderclap headaches
Amblyopia Cancer, liver Dermatofibrosarcoma protuberans Gout Lateral epicondylosis Osteosarcoma Retrograde ejaculation Thyroid cancer
Ameloblastoma Cancer, lung Dermatographia (Dermatographism) Granuloma annulare Latex allergy Otitis externa Rett syndrome Thyroid gland enlargement
Amenorrhea Cancer, male breast Dermatographism Granulomatosis with polyangiitis Lazy eye (amblyopia) Otitis media Reye's syndrome Thyroid nodules
American trypanosomiasis Cancer, Merkel cell Dermatomyositis Graves' disease LCIS Outer ear infection Rhabdomyosarcoma Thyroid, overactive
Amnesia Cancer, mesothelioma Dermatophytosis Grief Lead poisoning Ovarian cancer Rheumatic fever Thyroiditis, chronic lymphocytic
Amnesia, transient global Cancer, mouth Desmoid tumors Gross hematuria Left ventricular hypertrophy Ovarian cysts Rheumatoid arthritis TIA
Amnestic syndrome Cancer, multiple myeloma Desmoplastic small round cell tumors Group B strep disease Leg fracture Ovarian hyperstimulation syndrome Rhinitis, allergic Tic douloureux
Ampullary cancer Cancer, nasopharyngeal Detached retina Growing pains Legg-Calve-Perthes disease Overactive bladder Rhinitis, nonallergic Tinea barbae
Amyloid disease Cancer, neuroblastoma Developmental dysplasia of the hip Growth hormone excess Legionellosis Overactive thyroid Rhinitis, vasomotor Tinea capitis
Amyloidosis Cancer, non-Hodgkin's lymphoma Deviated septum Growth plate fractures Legionnaires' disease Oxyphil cell carcinoma Rib fracture Tinea corporis
Amyotrophic lateral sclerosis (ALS) Cancer, oral Devic's disease Guillain-Barre syndrome Leiomyosarcoma Pa Rickets Tinea cruris
Anal cancer Cancer, ovarian DFSP Gum disease, gingivitis Lentigines, solar PAD Rift Valley fever Tinea pedis
Anal fissure Cancer, pancreatic Diabetes Gum disease, periodontitis Leukemia PAES Ringing in the ear Tinea versicolor
Anal fistula Cancer, paraneoplastic syndromes Diabetes insipidus Gynecomastia Leukemia, acute lymphocytic Paget's disease of bone Ringworm (body) Tinnitus
Anal itching Cancer, prostate Diabetes mellitus H. pylori infection Leukemia, acute myelogenous Paget's disease of the breast Ringworm (scalp) TMD
Anaphylaxis Cancer, rectal Diabetes, gestational H1N1 flu Leukemia, chronic lymphocytic Pain, back Ringworm of the foot TMJ disorders
Anemia Cancer, retinoblastoma Diabetes, type 1 H1N1v flu Leukemia, chronic myelogenous Pain, chest RLS Toe walking in children
Anemia, aplastic Cancer, skin Diabetes, type 1 in children H3N2v flu Leukemia, general Pain, wrist Road rage Toenail fungus
Anemia, Cooley's Cancer, soft tissue sarcoma Diabetes, type 2 Hair loss Leukemia, hairy cell Painful bladder syndrome Rosacea Tongue cancer
Anemia, iron deficiency Cancer, squamous cell Diabetes, type 2 in children Hair-pulling disorder Leukoplakia Painful intercourse (dyspareunia) Roseola Tongue tie
Anemia, Mediterranean Cancer, stomach Diabetic coma Hairy cell leukemia Lewy body dementia Painful periods Rotator cuff injury Tongue-tie (ankyloglossia)
Anemia, sickle cell Cancer, testicular Diabetic hypoglycemia Hairy tongue Lice Palsy, Bell's Rotavirus Tonic-clonic (grand mal) seizure
Anemia, vitamin deficiency Cancer, throat Diabetic ketoacidosis Halitosis Lice, body Palsy, cerebral Round back Tonic-clonic seizure
Aneurysm, abdominal aortic Cancer, thyroid Diabetic nephropathy (kidney disease) Hammertoe and mallet toe Lichen planus Palsy, facial RSV Tonsil cancer
Aneurysm, aortic Cancer, uterine Diabetic neuropathy Hamstring injury Lichen planus, oral Pancreas inflammation Rubella Tonsillitis
Aneurysm, brain Cancer, vagina Diabetic retinopathy Hand fracture Lichen sclerosus Pancreatic cancer Rubeola Tooth abscess
Aneurysm, cerebral Cancer, vulvar Diaper rash Hand-foot-and-mouth disease Lichen simplex chronicus Pancreatic cysts Rumination syndrome Tooth decay
Aneurysm, popliteal Cancer, Wilms' tumor Diarrhea Hangovers Lip cancer Pancreatic neuroendocrine tumors Runner's knee Torn meniscus
Aneurysm, thoracic aortic Candidiasis, oral Diarrhea, antibiotic-associated Hantavirus pulmonary syndrome Lipodystrophy, intestinal Pancreatitis Ruptured disk Torsion, testicular
Aneurysms Candidiasis, vaginal Diarrhea, traveler's Hardening of the arteries Lipoma Panic attacks and panic disorder Ruptured eardrum (perforated eardrum) Torticollis, spasmodic
Angelman syndrome Canker sore Difficulty speaking Hashimoto's disease Liposarcoma PAPVR Ruptured spleen Total anomalous pulmonary venous connection
Angiitis Carbon monoxide poisoning Difficulty swallowing HAV infection Listeria infection Paraganglioma Sacral dimple Total anomalous pulmonary venous return (TAPVR)
Angina Carbuncles and boils DiGeorge syndrome (22q11.2 deletion syndrome)Hay fever Listeriosis Paraneoplastic neurological disorder Sacroiliitis Tourette syndrome
Angina pectoris Carcinoid syndrome Dilated cardiomyopathy HBV infection Liver cancer Paraneoplastic neurological syndrome SAD Toxemia
Angioedema and hives Carcinoid tumors Diphtheria HCC Liver disease Paraneoplastic syndromes of the nervous systemSalivary gland tumors Toxic epidermal necrolysis
Angiofollicular lymph node hyperplasia Carcinoma of unknown primary Dislocated shoulder HCV infection Liver failure, acute Parkinson's disease Salmonella infection Toxic hepatitis
Angiosarcoma Cardiac arrest, sudden Dissecting aneurysm Head and neck cancers Liver hemangioma Parotid tumors Salmonellosis Toxic shock syndrome
Ankle fracture Cardiac ischemia Dissociative disorders Head lice Liver spots Partial anomalous pulmonary venous connectionSarcoidosis Toxoplasmosis
Ankle sprain Cardiogenic shock Diverticulitis Head trauma, coma Liver, enlarged Partial anomalous pulmonary venous returnSarcoma Trachoma
Ankyloglossia Cardiomegaly Dizziness Headache, chronic daily Lobular carcinoma in situ (LCIS) Parvovirus infection Sarcoma, bone cancer Transient global amnesia
Ankylosing spondylitis Cardiomyopathy Double uterus Headache, cluster Localized scleroderma Patellar tendinitis Sarcoma, soft tissue Transient ischemic attack (TIA)
Anorexia Cardiomyopathy, dilated Double-outlet right ventricle Headache, exercise Lockjaw Patellofemoral pain syndrome SBBO Transposition of great vessels
Anorexia nervosa Cardiomyopathy, hypertrophic Down syndrome Headache, migraine Long QT syndrome Patent ductus arteriosus (PDA) Scabies Transposition of the great arteries
Anorgasmia in women Cardiospasm Dressler syndrome Headache, primary cough Loss of bladder control Patent foramen ovale SCAD Transverse myelitis
Anterior cruciate ligament injury Cardiovascular disease Drop foot Headache, sex Lou Gehrig's disease Pathological gambling Scalded mouth syndrome Traumatic brain injury
Anterior vaginal prolapse Carotid artery disease Drug addiction (substance use disorder) Headache, sinus Low blood pressure (hypotension) Pathological laughter and crying Scarlatina Traumatic grief
Anterior vaginal prolapse (cystocele) Carotid artery stenosis Drug allergy Headache, spinal Low blood sugar Pathological stealing Scarlet fever Traveler's diarrhea
Anthrax Carotid insufficiency Dry eyes Headache, tension Low body temperature PBA Schizoaffective disorder Trichinosis
Antibiotic-associated colitis Carpal tunnel syndrome Dry macular degeneration Headache, thunderclap Low platelet count PCOS Schizoid personality disorder Trichomoniasis
Antibiotic-associated diarrhea CAS Dry mouth Headaches in children Low sex drive in women PDA Schizophrenia Trichotillomania (hair-pulling disorder)
Antiphospholipid syndrome Castleman disease Dry skin Health anxiety Low sperm count Peanut allergy Schizophrenia, adolescent, childhood, juvenileTricuspid atresia
Antisocial personality disorder Cataracts Dry socket Hearing loss Low testosterone Pectus excavatum Schizotypal personality disorder Tricuspid valve disease
Anxiety disorder, generalized Cavernous malformations DSPS Heart arrhythmia Low-flow priapism Pediatric brain tumors Schwannoma Tricuspid valve regurgitation
Anxiety disorder, social Cavities and tooth decay DSRCT Heart attack Lung cancer Pediatric obstructive sleep apnea Schwannomatosis Trigeminal neuralgia
Anxiety disorders CDH DSWPD Heart defects in adults Lung collapse Pediatric white blood cell disorders SCI Trigger finger
Aortic aneurysm Celiac artery compression Ductal carcinoma in situ (DCIS) Heart disease Lung disease, interstitial Pediculosis capitis Sciatica Triple X syndrome
Aortic aneurysm, abdominal Celiac artery compression syndrome Dumping syndrome Heart failure Lupus Pelvic inflammatory disease (PID) Scleroderma Trisomy 21 syndrome
Aortic coarctation Celiac axis syndrome Dunbar syndrome Heart murmurs Lupus nephritis Pelvic organ prolapse Sclerosing cholangitis Trisomy X
Aortic dissection Celiac disease Dupuytren contracture Heart palpitations Lyme disease Pelvic pain, chronic Sclerosing mesenteritis Trouble swallowing
Aortic valve disease Celiac trunk compression syndrome Dural arteriovenous fistulas Heart valve disease Lymph nodes, swollen Pelvic support problems, uterine prolapseSclerosis, multiple Truncus arteriosus
Aortic valve regurgitation CELLO Dust mite allergy Heartburn Lymphadenitis Pemphigus Sclerosis, systemic Trypanosomiasis, American
Aortic valve stenosis Cellulite DVT Heat exhaustion Lymphadenitis, mesenteric Penicillin allergy Scoliosis Tuberculosis
APD Cellulitis Dwarfism Heat rash Lymphedema Peptic ulcer Scorpion sting Tuberous sclerosis
Aphasia Central auditory processing disorder Dysarthria Heatstroke Lymphoma Perforated eardrum Scrotal masses Tumor, salivary gland
Aphasia, primary progressive Central nervous system vascular malformationsDyshidrosis Heavy menstrual bleeding Lymphoma, Hodgkin's Pericardial effusion Seasonal affective disorder (SAD) Turner syndrome
Aphthous stomatitis Central sleep apnea Dyslexia Heel pain Lymphoma, non-Hodgkin's Pericardial inflammation Seasonal allergy Tympanic membrane perforation
Aphthous ulcer Cercarial dermatitis Dysmenorrhea Helicobacter pylori (H. pylori) infectionLymphoplasmacytic lymphoma Pericarditis Sebaceous carcinoma Type 1 diabetes
Apical ballooning syndrome Cerebral aneurysm Dyspareunia Hemangioma Lynch syndrome Perimenopause Sebaceous cysts Type 1 diabetes in children
Aplastic anemia Cerebral palsy Dyspepsia Hemangioma, liver Macular degeneration, dry Periodontal disease, gingivitis Seborrheic dermatitis Type 2 diabetes
Appendicitis Cerebrospinal fluid leak Dysphagia Hematoma, intracranial Macular degeneration, wet Periodontal disease, periodontitis Seborrheic keratosis Type 2 diabetes in children
APS Cerebrovascular accident Dysphonia Hematuria Mad cow disease Periodontitis Secondary brain cancer Typhoid fever
ARDS Cerumen impaction Dysrhythmias Hemifacial spasm Major depressive disorder Peripheral artery disease (PAD) Secondary hypertension Ulcer, aphthous
Arenaviruses Cervical cancer Dysthymia Hemochromatosis Malaria Peripheral nerve injuries SEID Ulcer, peptic
Argentine hemorrhagic fever Cervical dystonia Dystonia Hemolytic uremic syndrome (HUS) Male breast cancer Peripheral nerve tumors Seizure disorder Ulcer, rectal
Arm fracture Cervical osteoarthritis Dystonia, cervical Hemophilia Male hypogonadism Peripheral neuropathy Seizure, absence Ulcerative colitis
Arnold-Chiari malformation Cervical pain E. coli Hemorrhoids Male infertility Peripheral vascular disease Seizure, febrile Ulnar wrist pain
Arrhythmia Cervical spondylosis Ear infection (middle ear) Henoch-Schonlein purpura Malignant fibrous histiocytoma Peritonitis Seizure, grand mal Ulnar-sided wrist pain
Arteriosclerosis / atherosclerosis Cervicitis Ear infection, outer ear Hepatitis A virus infection Malignant hyperthermia Pernicious anemia Seizure, petit mal Umbilical hernia
Arteriovenous fistula Chagas disease Eardrum, ruptured Hepatitis B virus infection Malignant mesothelioma Pernio Seizure, temporal lobe Uncontrollable laughter and crying
Arteriovenous malformation Charcot-Marie-Tooth disease Early puberty Hepatitis C virus infection Malignant peripheral nerve sheath tumorsPeroneal muscular atrophy Seizures Under-eye puffiness
Arteritis, giant cell CHD Earwax blockage Hepatitis, autoimmune Mallet toe and hammertoe Persistent depressive disorder Selective IgA deficiency Underactive thyroid
Arteritis, Takayasu's Chemical dependency Eating disorders Hepatitis, toxic MALS Persistent post-concussive symptoms (Post-concussion syndrome)Self harming Undescended testicle
Arthritis Chemo brain Eating disorders, anorexia Hepatocellular carcinoma Mammary duct ectasia Personality disorder, antisocial Self-injury/cutting Undifferentiated pleomorphic sarcoma
Arthritis, basal joint Chest pain Eating disorders, binge eating Hepatolenticular degeneration Manic-depressive illness Personality disorder, borderline Separation anxiety disorder Upset stomach
Arthritis, degenerative Chest wall pain Eating disorders, bulimia Hepatomegaly Marfan syndrome Personality disorder, narcissistic Sepsis Ureteral cancer
Arthritis, gouty Chiari malformation Ebstein anomaly Hepatopulmonary syndrome MASLD Personality disorder, schizoid Septal defect, atrial Ureteral obstruction
Arthritis, infectious Chickenpox Ectopic heartbeat Hereditary hemochromatosis Mastalgia Personality disorder, schizotypal Septal defect, atrioventricular Urethral stricture
Arthritis, juvenile idiopathic Chilblains Ectopic pregnancy Hereditary hemorrhagic telangiectasia Mastitis Personality disorders Septal defect, ventricular Urinary incontinence
Arthritis, osteoarthritis Child maltreatment condition Ectropion Hereditary motor and sensory neuropathy MCAD deficiency Pertussis Septic arthritis Urinary tract infection (UTI)
Arthritis, psoriatic Childhood apraxia of speech Eczema Hereditary nonpolyposis colorectal cancer syndromeMD Pes planus Serotonin syndrome Urine color
Arthritis, reactive Childhood asthma Eczema, dyshidrotic Hernia, hiatal ME/CFS Pet allergy Severe acute respiratory syndrome (SARS)Urine, blood in
Arthritis, rheumatoid Childhood obesity ED Hernia, inguinal Measles Petit mal seizure Severe brain injury, coma Urticaria
Arthritis, septic Childhood schizophrenia Edema Hernia, umbilical Measles, German Peyronie disease Sex headaches Uterine cancer
Arthritis, thumb Chlamydia Edema, pulmonary Herniated disk Medial epicondylitis Pharyngitis Sexual dysfunction, female Uterine fibroids
Asbestosis Chlamydia trachomatis Egg allergy Herpes zoster Medial tibial stress syndrome Phenylketonuria (PKU) Sexual obsession Uterine leiomyoma
Ascariasis Cholangiocarcinoma (bile duct cancer) Ehlers-Danlos syndrome Herpes zoster oticus Median arcuate ligament syndrome (MALS) Pheochromocytoma Sexually transmitted diseases (STDs) Uterine myoma
ASD Cholangitis, primary sclerosing Ehrlichiosis and anaplasmosis Herpes, genital Medication overuse headaches Phlebitis Shaken baby syndrome Uterine polyps
Aseptic necrosis Cholecystitis Eisenmenger syndrome Hiatal hernia Mediterranean anemia Phobia, social Shellfish allergy Uterine prolapse
Aspergillosis Cholera Elevated blood pressure Hiccups Medulloblastoma Pick's disease Shigella infection Uterus didelphys
Asthma Cholestasis of pregnancy Embolism, pulmonary Hidradenitis suppurativa Megacolon PID Shin splints UTI
Asthma attack Chondrosarcoma Embryonal tumors High blood pressure (hypertension) Megaloblastic anemia Piles Shingles Uveitis
Asthma, childhood Chordoma Emphysema High blood pressure in children Melanoma Pilonidal cyst Short bowel syndrome V-fib
Asthma, exercise-induced Chorea, Huntington's Encephalitis High blood pressure, secondary Melanoma of the eye Pilonidal dimple Shoulder dislocation Vaginal agenesis
Asthma, occupational Choroid plexus carcinoma Encopresis High cholesterol MEN 2 Pimples Shy-Drager syndrome Vaginal atrophy
Astigmatism Chronic adrenal insufficiency End stage kidney failure High-flow priapism Meniere's disease Pinched nerve SIBO Vaginal cancer
Astrocytoma Chronic compartment syndrome End-stage renal disease Hilar cholangiocarcinoma Meningioma Pineoblastoma Sick sinus syndrome Vaginal candidiasis
AT Chronic cough Endocardial cushion defect Hip dysplasia Meningitis Pink eye (conjunctivitis) Sickle cell anemia Vaginal fistula
Ataxia Chronic daily headaches Endocarditis Hip fracture Meniscus tear Pinworm infection SIDS Vaginal prolapse, posterior
Atelectasis Chronic exertional compartment syndrome Endometrial cancer Hip labral tear Menopause Pituitary adenoma Sin Nombre virus infection Vaginitis
Atherosclerosis Chronic fatigue syndrome Endometrial polyps Hirschsprung's disease Menorrhagia Pituitary insufficiency Sinus headaches Valley fever
Athlete's foot Chronic granulomatous disease Endometriosis Hirsutism Menstrual cramps Pituitary tumors Sinusitis, acute Valvular heart disease
Atopic dermatitis (eczema) Chronic hives Enlarged breasts in men (gynecomastia) Histoplasmosis Mental illness Pityriasis rosea Sinusitis, chronic Variant CJD
Atrial fibrillation Chronic kidney disease Enlarged heart Hives and angioedema Meralgia paresthetica Pityriasis versicolor Sjogren's syndrome Varicella
Atrial flutter Chronic kidney failure Enlarged liver Hives, chronic Merkel cell carcinoma PKD SJS Varices, esophageal
Atrial septal defect (ASD) Chronic lymphocytic leukemia Enlarged prostate Hoarding disorder Mesenteric ischemia PKU Skin cancer Varicocele
Atrial tachycardia Chronic lymphocytic thyroiditis Enlarged spleen (splenomegaly) Hoarse voice Mesenteric lymphadenitis Placenta accreta Skin cancer, melanoma Varicose veins
Atrioventricular canal defect Chronic myelogenous leukemia Enlarged thyroid Hodgkin's disease Mesothelioma Placenta previa Skin writing Variola
Atrioventricular nodal reentry tachycardia (AVNRT)Chronic obstructive pulmonary disease Entropion Hodgkin's lymphoma (Hodgkin's disease) Metabolic dysfunction-associated steatotic liver diseasePlacental abruption Skin, itchy Vascular dementia
Atrioventricular septal defect Chronic pelvic pain Eosinophilic esophagitis Hordeolum Metabolic syndrome Plague Skipped heartbeats Vascular rings
Atrophic vaginitis Chronic renal failure Eosinophilic fasciitis Horner syndrome Metachromatic leukodystrophy Plantar fasciitis Slapped cheek disease Vasculitis
Attachment disorder Chronic sinusitis Eosinophilic granulomatosis with polyangiitisHorner-Bernard syndrome Metastatic prostate cancer Plantar neuroma Sleep apnea Vasomotor rhinitis
Attention-deficit/hyperactivity disorder (ADHD) in childrenChronic traumatic encephalopathy Ependymoma Hot flashes Metatarsalgia Plantar warts Sleep apnea in children, obstructive Vasovagal syncope
Attention-deficit/hyperactivity disorder in adultsChronic vulvar pain Epicondylitis, lateral HPV infection Methicillin-resistant staphylococcus aureus infectionPleurisy Sleep apnea, central vCJD
Atypical hyperplasia of the breast Churg-Strauss syndrome Epidermoid cysts Human immunodeficiency virus MG Pleuritis Sleep apnea, obstructive Velocardiofacial syndrome
Auditory information processing disorderCirrhosis Epidermolysis bullosa Human papillomavirus infection MGUS PLS Sleep disorders Venereal warts
Auditory processing disorder CJD Epididymitis Hunchback Microcephaly Pneumonia Sleep terrors Ventricular fibrillation
Autism spectrum disorder CKD Epiglottitis Hunt's syndrome Microscopic colitis Pneumonitis Sleep terrors (night terrors) Ventricular premature contraction
Autoimmune hepatitis Claudication Epilepsy Huntington's disease Microscopic hematuria Pneumothorax Sleepiness, daytime Ventricular septal defect (VSD)
Autoimmune pancreatitis Cleft lip and cleft palate Epilepsy, frontal lobe Hurthle cell cancer Migraine POEMS syndrome Sleeplessness Ventricular tachycardia
Autonomic neuropathy Click-murmur syndrome Epithelioid sarcoma Hutchinson-Gilford progeria syndrome Migraine with aura Poison ivy rash Sleepwalking Vertebral tumor
Avascular necrosis (osteonecrosis) Clinical depression Erectile dysfunction Hydrocele Mild cognitive impairment (MCI) Polio Slipped disk Vesicoureteral reflux
Avian influenza CLL Erythema infectiosum Hydrocephalus Miliaria Polycystic kidney disease Slow heartbeat Vestibular schwannoma
AVN Clostridioides difficile infection Escherichia coli infection Hydronephrosis Milk allergy Polycystic ovary syndrome (PCOS) Small bowel bacterial overgrowth VEXAS Syndrome
AVNRT Clostridium difficile infection Esophageal achalasia Hypercalcemia Milk intolerance Polycythemia vera Small bowel cancer Viral gastroenteritis (stomach flu)
Baby acne Clubfoot Esophageal cancer Hypercholesterolemia Miscarriage Polyhydramnios Small bowel prolapse Viral hemorrhagic fevers
Back pain Cluster headache Esophageal spasms Hypercholesterolemia, familial Mitral valve disease Polymorphous light eruption Small bowel prolapse (enterocele) Vitamin deficiency anemia
Bacterial vaginosis Coarctation of the aorta Esophageal varices Hypercortisolism Mitral valve prolapse Polymyalgia rheumatica Small intestinal bacterial overgrowth (SIBO)Vitiligo
Bad breath Coccidioidomycosis Esophagitis Hypereosinophilic syndrome Mitral valve regurgitation Polymyositis Small intestine prolapse Vocal cord paralysis
Bags under eyes Cold allergy Essential tremor Hyperglycemia in diabetes Mitral valve stenosis Polyps, colon Small vessel disease Vocal fold paralysis
Baker cyst Cold exposure Esthesioneuroblastoma Hyperhidrosis Mittelschmerz Polyps, endometrial Smallpox Voice disorders
Balance problems Cold sore Ewing sarcoma Hypermenorrhea Mixed connective tissue disease Polyps, nasal Snoring Von Willebrand disease
Baldness Cold urticaria Excess facial hair in women Hyperopia MOG antibody-associated disorder Polyps, stomach Social anxiety disorder (social phobia) VPCs
Barber's itch Cold, common Excess growth hormone Hyperoxaluria and oxalosis MOGAD Polyps, uterine Soft palate cancer Vulvar cancer
Barlow's syndrome Colic Excessive daytime sleepiness Hyperparathyroidism Molar pregnancy Pompholyx Soft tissue sarcoma Vulvodynia
Barotitis media Colitis, ischemic Excessive menstrual bleeding Hypersexuality Mold allergy Poor color vision Solar keratosis Waldenstrom macroglobulinemia
Barotrauma Colitis, microscopic Exercise headaches Hypertension Moles Popliteal artery aneurysm Solar lentigines Warts, common
Barrett's esophagus Colitis, pseudomembranous Exercise-induced asthma Hypertension, pregnancy-related Molluscum contagiosum Popliteal artery entrapment Solitary fibrous tumor Warts, genital
Bartholin's cyst Colitis, ulcerative External otitis Hypertension, pulmonary Mono Popliteal cyst Solitary rectal ulcer syndrome Warts, plantar
Basal cell carcinoma Collapsed lung Extrasystole Hyperthyroidism (overactive thyroid) Monoclonal gammopathy of undetermined significance (MGUS)Porphyria Somatic symptom disorder Warts, venereal
Basal joint arthritis Colon cancer Eye floaters Hypertrophic cardiomyopathy Mononucleosis Post acute sequelae of SARS COV-2 infection (PASC)Somnambulism Water on the knee
BDD Colon polyps Eye melanoma Hypoactive sexual desire disorder Mood disorders Post-cardiac injury syndrome Sore throat Wegener's granulomatosis
Bed-wetting Colonic ischemia Eyelid inflammation Hypochondria Morning sickness Post-chemotherapy cognitive impairment Spasmodic torticollis Wermer's syndrome
Bedbugs Color blindness Eyestrain Hypoglycemia Morphea Post-concussion syndrome Specific phobias Wet macular degeneration
Bedsores (pressure ulcers) Colorectal cancer Facial hair, excess in women Hypoglycemia, diabetic Morton's neuroma Post-polio syndrome Specific reading disability Wheat allergy
Bee sting Columnar epithelium lined lower esophagusFacial palsy Hypohydration Mosquito bites Post-traumatic stress disorder (PTSD) Spermatocele Whiplash
Behcet's disease Coma Factitious disorder Hyponatremia Mouth cancer Post-vasectomy pain syndrome Spider bites Whipple's disease
Bell's palsy Common cold Factor V Leiden Hypoparathyroidism Movement disorders Posterior cortical atrophy Spina bifida White blood cell disorders in children
Benign adrenal tumors Common cold in babies Fallen arches Hypopituitarism Moyamoya disease Posterior prolapse Spinal arteriovenous malformation (AVM) Whiteheads
Benign migratory glossitis Common variable immunodeficiency Fallot, tetralogy of Hypoplastic left heart syndrome MRSA infection Posterior vaginal prolapse (rectocele) Spinal cord injury Whooping cough
Benign paroxysmal positional vertigo Common warts Familial adenomatous polyposis Hypospadias MS Postherpetic neuralgia Spinal cord tumor Wilms tumor
Benign paroxysmal positional vertigo (BPPV)Complex regional pain syndrome Familial hypercholesterolemia Hypotension MSA Postmyocardial infarction syndrome Spinal curvature Wilson's disease
Benign peripheral nerve tumor Complicated bereavement Familial Mediterranean fever Hypothermia Mucocutaneous lymph node syndrome Postpartum depression Spinal headaches Wisdom teeth, impacted
Benign prostatic hyperplasia (BPH) Complicated grief Familial paroxysmal peritonitis Hypothyroidism (underactive thyroid) Multiple endocrine neoplasia type 1 Postpartum preeclampsia Spinal stenosis Wolff-Parkinson-White (WPW) syndrome
Berger's disease Compulsive gambling FAP IBD Multiple endocrine neoplasia, type 1 (MEN 1)Postural hypotension Splenomegaly Wolff-Parkinson-White accessory pathway-related tachycardia
Beta thalassemia Compulsive hoarding syndrome Farsightedness IBS Multiple endocrine neoplasia, type 2 (MEN 2)Pouchitis Spondylosis, cervical Wrinkles
Bicuspid aortic valve Compulsive overeating Febrile seizure Idiopathic hypersomnia Multiple myeloma Prader-Labhart-Willi syndrome Spontaneous coronary artery dissection (SCAD)Wrist fracture
Bile duct cancer Compulsive sexual behavior Fecal incontinence Idiopathic intracranial hypertension Multiple sclerosis Prader-Willi syndrome Sprained ankle Wrist pain
Bile reflux Compulsive stealing Female infertility Idiopathic myelofibrosis Multiple system atrophy Precocious puberty Sprains X-linked agammaglobulinemia
Binge drinking Concussion Female sexual dysfunction Idiopathic thrombocytopenic purpura Multisystem inflammatory syndrome in children (MIS-C)Prediabetes Squamous cell carcinoma of the skin Xerosis
Binge-eating disorder Condylomata acuminata Fetal alcohol syndrome Idiopathic toe walking Mumps Preeclampsia Stage 4 prostate cancer Xerostomia
Bipolar disorder Congenital adrenal hyperplasia Fetal macrosomia IgA nephropathy (Berger disease) Muscle cramp Preexcitation syndrome Stammering XXX syndrome
Bird flu (avian influenza) Congenital diaphragmatic hernia (CDH) Fever IgA vasculitis Muscle strains Pregnancy-related hypertension Staph infections Yeast infection (vaginal)
Black hairy tongue Congenital heart defects in children Fever blister IIH Muscular dystrophy Premature birth STDs Yersinia pestis
Blackheads Congenital heart disease in adults Fever, valley Illness anxiety disorder Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS)Premature ejaculation Steele-Richardson-Olszewski syndrome Yips
Bladder calculi Congenital hip dislocation Fevers, viral hemorrhagic Immune thrombocytopenia (ITP) Myasthenia gravis Premature ovarian failure Stein-Leventhal syndrome Zenker's diverticulum
Bladder cancer Congenital megacolon FH Immune thrombocytopenic purpura Myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD)Premature ovarian insufficiency Stenosing tenosynovitis Zollinger-Ellison syndrome
Bladder control, loss of Congenital mitral valve anomalies Fibroadenoma Impacted wisdom teeth Myelodysplastic syndromes Premature puberty Stenosis, aortic valve Zoster
Bladder exstrophy Congenital myasthenic syndromes Fibrocystic breasts Impaired glucose tolerance Myelofibrosis Premature ventricular contractions (PVCs)Stenosis, mitral valve
Bladder infection Congenital plagiocephaly Fibroids, uterine Impetigo Myocardial infarction Premenstrual syndrome Stenosis, pulmonary valve
Bladder inflammation Congestive heart failure Fibromuscular dysplasia Impotence Myocardial ischemia Premenstrual syndrome (PMS) Stenosis, pyloric
Bladder prolapse Conjoined twins Fibromyalgia Inborn error of metabolism Myocarditis Presbycusis Stenosis, spinal
Bladder stones Conjunctivitis Fibrosis, cystic Incompetent cervix Myoclonus Presbyopia Stevens-Johnson syndrome
Blastocystis hominis Constipation Fibrosis, interstitial pulmonary Incontinence, bowel Myofascial pain syndrome Prescription drug abuse Stings, bee
Convert the conversations to a ChatML SFT dataset
The conversation field is already a list of {role, content} ChatML turns (system/user/assistant), so conversion is just a projection:
import json
with open("opus5-5diseaseconversations.jsonl") as f, open("chatml.jsonl", "w") as w:
for line in f:
rec = json.loads(line)
# keep ONLY the conversation -> standard {"messages": [...]} SFT format
w.write(json.dumps({"messages": rec["conversation"]}) + "\n")
Or with the 🤗 datasets library:
from datasets import load_dataset
ds = load_dataset("nisten/opus5.5-doctor-patient-conversations-all-human-diseases", split="train")
chatml = ds.map(lambda r: {"messages": r["conversation"]}, remove_columns=ds.column_names)
chatml.to_json("chatml.jsonl")
Build a simple RAG over it
Each conversation averages ~5,700 tokens (median ~5.7k, p95 ~7.0k, max ~8.4k) — longer than the Opus 4.8 set, so budget accordingly. Do not embed full conversations — index a compact retrieval key per disease and keep the full conversation as the payload you feed the LLM. This keeps embeddings clean and lets you budget context by the known average length.
import json, numpy as np
from sentence_transformers import SentenceTransformer
recs = [json.loads(l) for l in open("opus5-5diseaseconversations.jsonl")]
model = SentenceTransformer("BAAI/bge-small-en-v1.5")
# 1) Embed a COMPACT key (name + aliases + description + summary ≈ a few hundred tokens),
# NOT the ~5.7k-token conversation.
def key(r):
return f"{r['name']}. {' '.join(r['aliases'])}. {r['description']} {r['executive_summary']}"
emb = model.encode([key(r) for r in recs], normalize_embeddings=True) # (2194, d)
def search(query, k):
qv = model.encode([query], normalize_embeddings=True)[0]
idx = (emb @ qv).argsort()[::-1][:k] # cosine sim, top-k
return [recs[i] for i in idx]
# 2) Budget k by the avg conversation length so retrieved context fits your model window.
AVG_CONVO_TOK = 5700
def build_context(query, model_window=32000, reserve_for_prompt=4000):
k = max(1, (model_window - reserve_for_prompt) // AVG_CONVO_TOK) # 32k -> k=4
hits = search(query, k)
convo = lambda r: "\n".join(f"{m['role']}: {m['content']}" for m in r["conversation"])
return "\n\n---\n\n".join(convo(r) for r in hits)
# context = build_context("35 pack-year smoker, hemoptysis, weight loss, lung mass")
# -> feed `context` + the user question to your chat model
Rough k guide: 8k window → k=1 (tight; pass the executive_summary instead); 16k → k=2; 32k → k≤4; 128k → k≤21. If you need more breadth at small context, retrieve on the compact key but pass each hit's executive_summary + structured fields instead of the full conversation.
Generation & verification
Generated and verified agentically with Claude Opus 5.5, one disease per agent. One orchestrator merged the first pass; the six retry records merged under a file lock, and the file was re-checked for duplicates and torn lines (none). Demographics were sampled per record from weighted workforce/population tables so clinician and patient ethnicities differ and reflect realistic distributions. High-consequence topics (select agents, pandemic pathogens) were produced under rigorous clinical-ID framing and held to the same bar as routine diseases.
Every record had to pass, at commit time: the full 20-key schema; strict system → user/assistant alternation; >=15 turns and >=8,000 characters; >=3 items in each list field; zero emoji; no filler openers; no leaked generation instructions; vitals and labs with units and reference ranges that agree with each other (ABG vs Henderson-Hasselbalch, anion gap vs chemistry, doses vs weight); drugs taken from the record's own related_drugs; one tactfully caught patient non-disclosure; and every PMID in the text resolved live via NCBI E-utilities with its title matched to what the model claimed. Records that failed went back to the agent, up to three times, then were dropped and regenerated by a fresh agent.
After generation the whole file was re-audited: a seeded 10-record clinical audit (accuracy, numerical consistency, drug facts, citation integrity, register, pedagogy, safety), a full re-run of every offline gate over all 2,194 records (zero failures), and the pubmed_refs rebuild described under Provenance. Corrections made after generation, all logged: one citation to a retracted paper replaced, the two source-text fixes above, and fifteen patient_scenario fields stripped of a sentence that echoed the demographic-roll instruction.
Agentic traces are intentionally excluded.
Limitations & disclaimer
Synthetic data for research, RAG grounding, and model training only. Not medical advice; do not use for diagnosis or treatment. Each conversation contains exactly one deliberate patient non-disclosure (an intentional teaching feature) that the clinician detects and corrects. PMIDs were verified to exist and to carry the title the model claimed — that does not guarantee the paper says what the conversation says it says. Verify any clinical fact against primary sources before downstream use.
Citation
@misc{doctor_patient_all_human_diseases_opus55,
title = {Doctor-Patient Conversations — All Human Diseases (Opus 5.5)},
year = {2026},
note = {Synthetic clinical dialogue over 2,194 human diseases with NCBI-verified citations; sequel to nisten/opus-doctor-patient-conversations-all-human-diseases; seed disease list derived from nisten/all-human-diseases},
url = {https://huggingface.co/datasets/nisten/opus5.5-doctor-patient-conversations-all-human-diseases}
}
License: MIT.
Sample record
One record from this Opus 5.5 set with every conversation turn included; the three list fields are cut to their first item here and the verbatim source fields are omitted. In the file it is a single compact JSON line; here the keys are shown as code and the long text values as plain markdown so it word-wraps on this page. Verbatim source fields plus the agentically generated clinical encounter — diabetic ketoacidosis from ED arrival through resolution, with Dr. Anahit Sarkissian, an endocrinologist in Pasadena, managing Jiayi Chen, a 26-year-old with type 1 diabetes who initially overstates her basal-insulin adherence. The ABG, anion gap and insulin-infusion numbers all agree with each other, and the three pubmed_refs are exactly the papers the conversation cites.
{
"name": "Diabetic ketoacidosis"
"icd10": "E10.10"
"body_systems": ["endocrine","renal"]
(verbatim source fields omitted here for brevity: aliases, search, description, related_drugs, drug_interactions, food_interactions, prevalence, source_disease)
"pubmed_refs": [
{"pmid":"39052901","year":2024}PMID 39052901 — Hyperglycemic Crises in Adults With Diabetes: A Consensus Report.{"pmid":"19564476","year":2009}PMID 19564476 — Hyperglycemic crises in adult patients with diabetes.{"pmid":"26078479","year":2015}PMID 26078479 — Euglycemic Diabetic Ketoacidosis: A Potential Complication of Treatment With Sodium-Glucose Cotransporter 2 Inhibition.]
"clinician_persona": Dr. Anahit Sarkissian, MD, is an Armenian-American internist and board-certified endocrinologist who attends on the inpatient diabetes and medical step-down service at an academic medical center in Pasadena, California, where she runs the hospital's hyperglycemic-crisis protocol and teaches residents insulin-infusion management.
"patient_scenario": Jiayi Chen is a 26-year-old Chinese-American woman, a UX researcher at a small software startup, with type 1 diabetes since age 11 on multiple daily injections (insulin glargine 18 units nightly plus insulin lispro with meals). She aged off her parents' insurance in June, moved to a high-deductible employer plan, and presents to the emergency department at 01:40 with 36 hours of polyuria, polydipsia, vomiting, diffuse abdominal pain, and air hunger. She initially reports perfect basal-insulin adherence; her pen later shows she has been taking less than half her prescribed glargine dose.
"conversation": [ (33 turns — patient turns quoted, clinician turns plain)
{"role":"system"}
You are Dr. Anahit Sarkissian, an internist and endocrinologist attending on the inpatient diabetes and medical step-down service at an academic medical center in Pasadena, California. You are managing an adult with a hyperglycemic crisis from emergency department arrival through resolution and discharge planning. Communicate with precision, show your arithmetic when numbers drive decisions, cite evidence where it matters, and keep patient safety and shared decision-making central. Avoid emojis and filler.
{"role":"user"}
I'm Jiayi Chen, I'm 26. I've had type 1 diabetes since I was eleven. I've been throwing up since yesterday afternoon, maybe six or seven times, and my whole stomach hurts. I can't stop drinking water and I'm peeing constantly. My meter just said HI at home, so my roommate drove me in. I feel like I can't get a full breath.
{"role":"assistant"}
Jiayi, I'm Dr. Sarkissian, the endocrinology attending tonight. Vomiting, diffuse abdominal pain, heavy thirst and urination, a meter reading of HI, and air hunger in someone with type 1 diabetes point strongly toward diabetic ketoacidosis (DKA). We are treating this as an emergency, and several things are happening in parallel while we talk.
Your triage numbers:
- BP 102/64 mmHg, HR 118 bpm, RR 28 breaths/min (deep and regular), SpO2 99% on room air, temperature 37.4 °C
- Weight 58 kg (you usually weigh about 61 kg), height 163 cm
- Point-of-care glucose 486 mg/dL; capillary beta-hydroxybutyrate 6.2 mmol/L (normal below 0.6)
The deep breathing is your lungs blowing off carbon dioxide to offset acid in your blood. It is called Kussmaul respiration, and your breath has the fruity smell of acetone, which is a ketone.
Right now the team is placing two IV lines, drawing a chemistry panel, blood gas, serum ketones, lactate, CBC, lipase, troponin and a pregnancy test, running an ECG to look for potassium effects on the heart, and starting the first liter of 0.9% sodium chloride.
A few quick questions; short answers are fine:
- When did you last take your long-acting insulin, and what dose?
- Any fever, cough, burning with urination, or chest pain?
- Any new medicines, alcohol, or cannabis in the past week?
- Is there any chance you could be pregnant?
{"role":"user"}
Lantus, eighteen units every night. I haven't missed any. I use Humalog for meals but I haven't eaten since yesterday so I skipped those. No fever. I had a scratchy throat last week, that's all. No alcohol, no weed. I'm not pregnant, I have an IUD. Why does my stomach hurt so much if it's my sugar?
{"role":"assistant"}
Abdominal pain is common in DKA and tracks with the severity of the acidosis rather than with the glucose; the ketoacids and delayed gastric emptying both contribute. It usually settles as the acid clears. The trap is assuming that, because a surgical problem such as appendicitis or pancreatitis can also trigger DKA. So I examine you now and again once the acidosis improves; pain that persists after correction gets imaging.
My exam:
- Dry mucous membranes, capillary refill 3 seconds, flat neck veins: significant volume depletion
- Abdomen soft, diffusely tender, no guarding or rebound, hypoactive bowel sounds
- Lungs clear; no rash; alert and oriented, GCS 15, though you are clearly exhausted
We will run the pregnancy test even with an IUD in place. IUDs are very effective, but if a pregnancy does occur with one, it has a higher chance of being ectopic, and pregnancy changes how DKA behaves and how we treat it. It is routine for every woman of reproductive age with DKA.
The scratchy throat is worth noting but a mild viral illness rarely produces ketoacidosis this deep by itself when basal insulin is on board. I will come back to that once the labs return.
{"role":"user"}
Okay. The nurse just said my potassium is high. Isn't high potassium dangerous?
{"role":"assistant"}
Your first labs are back, so let me walk through them, because the potassium only makes sense in context.
Chemistry
- Sodium 131 mmol/L (135–145)
- Potassium 5.4 mmol/L (3.5–5.0)
- Chloride 96 mmol/L (98–107)
- Bicarbonate 8 mmol/L (22–29)
- BUN 28 mg/dL (7–20), creatinine 1.3 mg/dL (0.5–1.1)
- Glucose 486 mg/dL (70–99 fasting)
- Serum beta-hydroxybutyrate 6.8 mmol/L (<0.6), lactate 1.4 mmol/L (0.5–2.0)
- Albumin 4.4 g/dL (3.5–5.0), phosphate 4.6 mg/dL (2.5–4.5), magnesium 1.9 mg/dL (1.7–2.2)
Arterial blood gas (room air): pH 7.20 (7.35–7.45), PaCO2 21 mmHg (35–45), PaO2 112 mmHg (80–100), HCO3 8 mmol/L
Other: WBC 16.8 x10^9/L (4.0–11.0), 86% neutrophils; hemoglobin 14.6 g/dL (12.0–15.5, hemoconcentrated); lipase 88 U/L (13–60); high-sensitivity troponin I <5 ng/L (<16); serum hCG negative; urinalysis glucose >1000 mg/dL, ketones large, nitrite and leukocyte esterase negative. ECG: sinus tachycardia at 118 bpm, no peaked T waves, QTc 438 ms.
The arithmetic
- Anion gap = 131 − (96 + 8) = 27 mmol/L (normal about 8–12). Albumin is normal, so no correction is needed.
- Henderson-Hasselbalch check: 6.1 + log10(8 / (0.03 x 21)) = 6.1 + log10(12.7) = 6.1 + 1.10 = 7.20, matching the measured pH.
- Winter's formula: expected PaCO2 = 1.5 x 8 + 8 = 20 ± 2 mmHg. Your 21 mmHg means your breathing is compensating appropriately, with no separate respiratory problem.
- Delta ratio = (27 − 12) / (24 − 8) = 15/16 = 0.94: a pure anion-gap acidosis.
- Corrected sodium = 131 + 1.6 x (486 − 100)/100 = about 137 mmol/L. The low measured sodium is dilutional from glucose pulling water out of cells.
- Calculated osmolality = 2 x 131 + 486/18 + 28/2.8 = 262 + 27 + 10 = 299 mOsm/kg (275–295).
This meets every criterion for DKA in the 2024 consensus report: hyperglycemia in known diabetes, beta-hydroxybutyrate at or above 3.0 mmol/L, and pH below 7.3 or bicarbonate below 18 mmol/L (PMID: 39052901). A bicarbonate below 10 mmol/L puts you in the severe category, so you will go to the step-down unit with hourly monitoring.
Now the potassium. The 5.4 is misleading. Your body as a whole is short of potassium, typically 3 to 5 mEq per kg in DKA, which for you is roughly 174 to 290 mEq, lost through all that urination and vomiting. The blood level reads high because acidosis and the absence of insulin push potassium out of cells. Once insulin starts, it drives potassium back into cells and the blood level can fall fast. So the danger here is actually the drop that is coming. Because you are above 5.0 with a normal ECG, the first bags run without potassium and we recheck every 2 hours, adding potassium as soon as you fall below 5.0. If you had come in below 3.5 mmol/L, I would hold insulin entirely until potassium was replaced.
The mild kidney changes (BUN-to-creatinine ratio about 22) reflect dehydration. Estimated creatinine clearance by Cockcroft-Gault is (140 − 26) x 58 / (72 x 1.3) x 0.85 = about 60 mL/min, and it should recover with fluids. Acetoacetate can also falsely raise creatinine on some lab assays.
{"role":"user"}
So what's the actual treatment? Just insulin?
{"role":"assistant"}
Insulin is essential but it comes second in a set of three, plus a search for the trigger.
1. Fluids: 0.9% sodium chloride (normal saline)
- First liter over the first hour, about 17 mL/kg, now running.
- Then 250 mL/h, about 4.3 mL/kg/h, adjusted to blood pressure, heart rate and urine output.
- Your total water deficit is probably near 100 mL/kg, roughly 5.8 L, replaced over 24 to 48 hours rather than all at once.
2. Insulin: regular insulin (Humulin R) by continuous IV infusion
- Bolus 0.1 units/kg = 5.8 units IV, then 0.1 units/kg/h = 5.8 units/h.
- The bag is 100 units in 100 mL of normal saline, so 1 unit/mL and the pump runs at 5.8 mL/h. The tubing is primed with insulin first because insulin sticks to plastic.
- Target glucose fall: 50 to 75 mg/dL per hour. Fingerstick glucose hourly, electrolytes every 2 hours, beta-hydroxybutyrate every 2 to 4 hours.
- When glucose drops below 250 mg/dL, we add dextrose to your fluids and halve the infusion to 0.05 units/kg/h = 2.9 units/h. The goal is to clear the ketones, and ketones clear more slowly than sugar. Stopping insulin because the glucose looks good is one of the classic ways DKA relapses.
3. Potassium chloride IV
- Held while potassium is above 5.0; then 20 to 30 mEq per liter of fluid to keep potassium between 4.0 and 5.0 mmol/L.
4. Find the precipitant: infection, insulin interruption, new medicines, pancreatitis, a cardiac event, or pregnancy.
This framework comes from the ADA consensus on hyperglycemic crises (PMID: 19564476), with adult mortality below 1% when it is followed, and it was updated in the 2024 multisociety report (PMID: 39052901). The report notes that balanced crystalloids are a reasonable alternative to saline; saline is what we are using, and I will watch your chloride because of it.
{"role":"user"}
My sister is texting me that I should be getting bicarbonate since my blood is acidic. She's a nurse in Toronto. Should I?
{"role":"assistant"}
It is a reasonable thing to ask, and your sister is right that bicarbonate exists for this situation, but only at a much more extreme level than yours.
Where the threshold sits
- The 2009 ADA consensus reserved sodium bicarbonate for pH below 6.9, typically 100 mmol in 400 mL of sterile water with 20 mEq potassium chloride over 2 hours (PMID: 19564476).
- The 2024 update suggests considering it only when pH is below 7.0 (PMID: 39052901).
- Your pH is 7.20, with a bicarbonate of 8 mmol/L and good respiratory compensation.
Why giving it now would do more harm than good
- Potassium: bicarbonate pushes potassium into cells, on top of what insulin is already doing. With a total-body deficit of roughly 174 to 290 mEq, that could drop you into dangerous hypokalemia and arrhythmia risk.
- Paradoxical acidosis in the brain: the CO2 generated crosses into the cerebrospinal fluid faster than bicarbonate does.
- Slower ketone clearance: studies have associated bicarbonate with delayed fall in ketones.
- A sodium and volume load, and in children an association with cerebral edema.
How your acid will clear instead: insulin switches off ketone production, and the ketoacids you already have are metabolized, which regenerates bicarbonate. That is why your bicarbonate should rise steadily over the next 12 to 18 hours without us giving any.
If your pH fell below 7.0 or your blood pressure failed to respond to fluids, I would revisit it. Please tell your sister we are watching exactly that.
{"role":"user"}
It's been a few hours. I stopped throwing up, but I'm still so thirsty. The nurse keeps poking my finger every hour. What are the numbers now?
{"role":"assistant"}
You are responding the way we want. Here is the trend at hour 3.
Vitals: BP 110/68 mmHg, HR 104 bpm, RR 24/min, urine output about 120 mL/h.
Hourly glucose (mg/dL): 486, 441, 389, 338. That is a fall of about 50 mg/dL per hour, at the bottom of our 50 to 75 target, which is where I want it: steady rather than dramatic.
Hour 2 chemistry
- Sodium 133, potassium 4.7, chloride 101, bicarbonate 10 mmol/L
- Anion gap = 133 − (101 + 10) = 22, down from 27
- Beta-hydroxybutyrate 5.1 mmol/L, down from 6.8
- BUN 26 mg/dL, creatinine 1.2 mg/dL
- Corrected sodium = 133 + 1.6 x (389 − 100)/100 = about 137.6, essentially unchanged from 137. The measured sodium rising as glucose falls is expected. A rapidly falling corrected sodium would concern me because of cerebral edema risk, and yours is stable.
Potassium: at 4.7 you crossed below 5.0, so potassium chloride 20 mEq per liter is now in your saline. At 250 mL/h that is 5 mEq per hour, a safe peripheral rate.
Fluids so far: 1,000 mL in the first hour plus 2 hours at 250 mL/h = 1.5 L.
The thirst is from your blood still being concentrated; it will ease as we replace volume. Small sips of water or ice chips are fine now that the vomiting has stopped.
{"role":"user"}
Why did this even happen? I've been doing everything right.
{"role":"assistant"}
Finding the cause matters as much as the treatment, because it decides whether this happens again. The usual precipitants are infection, missed or reduced insulin, insulin that has lost potency, new medications (steroids, some antipsychotics, SGLT2 inhibitors, sometimes thiazide diuretics), pancreatitis, a heart attack, or pregnancy.
What your workup shows so far:
- No fever, clear lungs, and a clean urinalysis.
- WBC 16.8 x10^9/L: leukocytosis is common in DKA from stress hormones alone. Counts above about 25 would push me harder to look for infection.
- Lipase 88 U/L: nonspecific elevations are common in DKA, and pancreatitis generally needs more than 3 times the upper limit, which here is 180 U/L, plus a compatible exam.
- Troponin negative, ECG without ischemia, hCG negative.
- A mild sore throat a week ago that has already resolved.
That leaves insulin as the most likely explanation. A beta-hydroxybutyrate of 6.8 mmol/L tells me your body has had very little effective insulin for at least a day or two, and probably low levels for longer than that. That can happen even when someone is injecting faithfully: a pen left in a hot car in a Pasadena September can lose potency, a needle can clog, or an injection site with scar tissue can absorb poorly.
Have you noticed anything like that? Pens left in heat, doses that seemed not to work, anything that changed with your insulin supply in the past few months?
{"role":"user"}
The pens are fine. I keep them in my bag. I told you, I take it every night.
{"role":"assistant"}
Understood, and I'm not doubting that you inject every night. Would you be comfortable letting me look at the pen in your bag with you? Two practical reasons: heat-damaged insulin can look perfectly clear, and before we restart glargine later I want to see the pen, its open date, and your needles. If the pen is the problem, I would rather find it now than send you home with it.
{"role":"user"}
Fine. It's in the front pocket. The one with the date written on the label.
{"role":"assistant"}
Thank you. The pen is clear and was stored at room temperature, which is good. I want to talk through what I see, because the numbers do not add up the way you described, and I would rather ask you directly than guess.
- The label says it was opened on August 30. Today is September 20, so that is 21 days.
- A Lantus SoloStar pen holds 300 units.
- At 18 units a night plus the 2-unit safety test before each dose, you would use about 20 units per day, so a full pen should run out in about 15 days.
- The scale on the cartridge shows roughly 120 units still left. That means about 180 units were used over 21 days, around 8.6 units per day, less than half of your prescribed 18.
I'm not raising this to catch you out. A large number of the young adults I treat for DKA have been stretching their insulin, and it is almost always about cost or access, not carelessness. If that is what has been going on, it is the single most useful thing you could tell me, because it is fixable and it is exactly what put you here tonight.
{"role":"user"}
Okay. I turned 26 in June and got kicked off my parents' insurance. My company's plan has a huge deductible and the pharmacy wanted 340 dollars for a box of pens. I started doing about half the Lantus so it would last, and I skipped some Humalog too. I didn't want my parents to know I couldn't afford it. I'm sorry I said I hadn't missed any.
{"role":"assistant"}
Thank you for telling me. You don't owe me an apology; you did something a lot of people do when a system prices them out of a drug they cannot live without. It also explains tonight completely: halving basal insulin in type 1 diabetes leaves the liver producing ketones around the clock, and a minor viral illness on top of that was enough to tip you over.
Here is what changes:
Before you leave
- Our diabetes pharmacist and social worker will see you this morning. Manufacturer savings programs, including the one for Lantus, cap out-of-pocket cost at about 35 dollars a month for many commercially insured and uninsured patients; eligibility varies by plan, and they will enroll you before discharge.
- They will check your plan's formulary. An equivalent basal insulin on the preferred tier may cost far less even before the deductible.
- You go home with insulin in hand, not a prescription you cannot fill.
One non-negotiable rule: basal insulin is never the dose you cut. If money or supply ever gets tight again, call our clinic before you reduce anything. Human insulins such as Humulin R and NPH are sold without a prescription at lower cost in the US, but they work very differently from Lantus and Humalog, and switching without a written dosing plan from us can cause severe lows or another episode like this one.
Whether you tell your parents is your decision, and nothing about your care depends on it.
{"role":"user"}
It's morning now I think. I'm less foggy. The nurse just switched my IV bag to one that has sugar in it. Doesn't that defeat the purpose?
{"role":"assistant"}
It is the step that lets us keep treating the ketones safely. At hour 6 your glucose reached 236 mg/dL, below our 250 threshold, but the acid is only partly cleared:
Hour 6 chemistry
- Sodium 136, potassium 4.0, chloride 105, bicarbonate 13 mmol/L
- Anion gap = 136 − (105 + 13) = 18, still above normal
- Beta-hydroxybutyrate 3.4 mmol/L, halved from 6.8 but still high
- BUN 21 mg/dL, creatinine 1.0 mg/dL; corrected sodium = 136 + 1.6 x (236 − 100)/100 = about 138
Ketones take longer to clear than glucose. If we let the glucose keep falling, we would have to stop the insulin to avoid hypoglycemia, and the ketones would come right back. So instead:
- Fluid changed to 5% dextrose in 0.45% sodium chloride with potassium chloride 20 mEq/L at 200 mL/h. The half-strength saline is appropriate because your corrected sodium is normal and you still need free water.
- Insulin reduced to 0.05 units/kg/h = 2.9 units/h, per the 2024 consensus (PMID: 39052901).
- Target glucose 150 to 250 mg/dL while the insulin keeps shutting down ketone production.
Potassium at 4.0 is in our 4.0 to 5.0 target, so the replacement stays at 20 mEq/L for now, with a recheck in 2 hours. Your heart rate is 96 bpm and BP 116/72 mmHg. Fluids through hour 6 total about 2.25 L.
{"role":"user"}
Can I eat something? My mom is driving down from Arcadia with jook, rice porridge. And when can I go home?
{"role":"assistant"}
Your nausea is gone and your abdomen is much softer on exam, so yes, you can start eating. Plain rice porridge is a very good first meal after DKA: gentle, hydrating, and easy to estimate. A cup of plain rice congee typically runs roughly 20 to 30 grams of carbohydrate, depending on how thick it is. While the insulin drip is running, the nurses will simply check your glucose and adjust; once you are back on injections, you will cover it with Humalog using your usual ratio.
For going home, I need all of the following:
- DKA resolved by the 2024 criteria: beta-hydroxybutyrate below 0.6 mmol/L and either venous pH at least 7.3 or bicarbonate at least 18 mmol/L (PMID: 39052901).
- Eating and drinking without vomiting.
- Back on subcutaneous insulin with a safe overlap, and glucose stable for a full day on it.
- Insulin you can actually afford, physically in your hands.
- Sick-day and ketone-testing teaching done.
If the next 12 hours go the way the first six have, that most likely means discharge tomorrow.
{"role":"user"}
The nurse said my bicarbonate is still low but the gap is almost normal? I don't understand how both can be true.
{"role":"assistant"}
Both are true, and the reason is the saline we gave you, not the ketones.
Hour 12 results
- Glucose 184 mg/dL
- Sodium 138, potassium 3.9, chloride 108, bicarbonate 16 mmol/L
- Anion gap = 138 − (108 + 16) = 14, near normal
- Beta-hydroxybutyrate 1.2 mmol/L
- Venous gas: pH 7.32, PvCO2 32 mmHg, HCO3 16. Check: 6.1 + log10(16 / (0.03 x 32)) = 6.1 + log10(16.7) = 6.1 + 1.22 = 7.32.
- Phosphate 1.9 mg/dL (2.5–4.5), magnesium 1.8 mg/dL, creatinine 0.9 mg/dL
Reading them together
- The anion gap has closed from 27 to 14, an excess of only 2 over normal. The bicarbonate is still down by 24 − 16 = 8. If ketoacids were the only thing lowering your bicarbonate, those two would move in step.
- The difference is chloride, which rose from 96 to 108 mmol/L. Normal saline contains 154 mmol/L of chloride, far more than blood, and several liters of it produce a mild hyperchloremic, non-anion-gap acidosis. It is harmless and resolves over a day or two as your kidneys excrete the chloride.
- This is why I follow beta-hydroxybutyrate and the anion gap rather than bicarbonate alone. Chasing the bicarbonate here would keep you on an insulin drip for no benefit.
Potassium at 3.9 has dipped below 4.0, so replacement goes up to 30 mEq per liter. Phosphate at 1.9 is expected as insulin drives it into cells. Replacement has not shown benefit unless phosphate falls below 1.0 mg/dL or there is heart or breathing muscle weakness (PMID: 19564476), so we will recheck it with the next panel rather than treat it.
{"role":"user"}
My mom brought the jook and I kept a whole bowl down. Are we done with the drip?
{"role":"assistant"}
Almost. Your hour 16 labs meet resolution criteria:
- Beta-hydroxybutyrate 0.4 mmol/L (below 0.6)
- Bicarbonate 19 mmol/L, venous pH 7.36, PvCO2 35 mmHg. Check: 6.1 + log10(19 / (0.03 x 35)) = 6.1 + log10(18.1) = 6.1 + 1.26 = 7.36.
- Sodium 139, chloride 109, anion gap = 139 − (109 + 19) = 11
- Potassium 4.0 mmol/L, glucose 172 mg/dL
- HR 84 bpm, BP 118/74 mmHg, RR 16/min, eating and drinking
The transition, which is where many relapses happen
- IV regular insulin has a half-life of only a few minutes. If the drip stops before long-acting insulin is working, ketone production restarts within hours.
- So you receive insulin glargine (Lantus) 18 units subcutaneously now, your full prescribed dose, about 0.31 units/kg. Glargine takes roughly 1 to 2 hours to begin acting, so the drip continues for 2 more hours and then stops.
- From your next meal, Humalog at your usual 1 unit per 10 grams of carbohydrate plus correction. A bowl of jook at about 25 grams would be about 2.5 units.
- Fingerstick glucose before meals and at bedtime overnight.
On your home dose: your A1c came back at 10.4% (below 5.7% normal); it was 7.6% in March. That rise reflects the months of half-dose basal rather than a dose that was wrong. Your current total daily insulin of roughly 34 units, about 0.59 units/kg, is appropriate for your weight, so we restart it as prescribed and adjust in clinic.
Fluids totaled about 4.25 L over 16 hours. Potassium given so far is roughly 68 mEq, well short of the estimated deficit; eating will make up most of the rest, and we will give oral potassium chloride if your level falls below 3.5 mmol/L.
{"role":"user"}
Before I go, a coworker with type 2 takes Jardiance and lost weight on it. A friend said some type 1 people take it off-label so they need less insulin. Could I try that? Especially if it means buying less insulin.
{"role":"assistant"}
For you specifically, I would strongly advise against it, and this is the most important medication warning I will give you.
SGLT2 inhibitors and insulin: a major interaction for type 1 diabetes
- Empagliflozin (Jardiance) and the related drugs make the kidneys excrete glucose. That lowers blood sugar without the insulin needed to stop ketone production, and they also raise glucagon and cause volume loss.
- The result is euglycemic DKA: ketoacidosis with a glucose that can be close to normal, sometimes under 200 or even 150 mg/dL. You would feel sick and see a reassuring meter reading, and so might an emergency physician.
- In the case series that first described this, Peters and colleagues identified 13 episodes of euglycemic DKA or ketosis in 9 patients, 7 of whom had type 1 diabetes, and the absence of significant hyperglycemia delayed recognition by both patients and clinicians (PMID: 26078479). They concluded these drugs should be used in type 1 diabetes only with great caution, extensive counseling, and close monitoring.
- The riskiest combination is an SGLT2 inhibitor plus reduced insulin, which is exactly the situation you just came from. SGLT2 inhibitors are not FDA-approved for type 1 diabetes in the United States.
Thiazide diuretics: if anyone ever proposes hydrochlorothiazide or chlorthalidone, for example for blood pressure years from now, remind them you have type 1 diabetes. Thiazides can raise glucose and worsen insulin resistance, which increases insulin needs and has precipitated DKA. They also lower potassium. They are not forbidden, but they need a plan and closer glucose checks.
The other common culprit is steroids such as prednisone or a steroid injection; any course needs a temporary insulin increase.
If cost is the concern, the answer is the savings program and formulary work the pharmacist is doing, not a drug that removes your safety margin.
{"role":"user"}
So what do I do if I get sick again, or my sugar goes high at home?
{"role":"assistant"}
This is your sick-day plan. You will get it in writing, along with a blood ketone meter and strips.
Rule 1: never stop or cut your Lantus, even if you are not eating. Basal insulin is what stops ketone production. Mealtime Humalog is what you adjust to food.
Rule 2: test more often. When you are ill, check glucose every 2 to 4 hours, and check blood ketones whenever glucose is above 250 mg/dL on two checks in a row, or any time you have nausea, vomiting or abdominal pain, regardless of glucose.
Rule 3: act on the ketone number.
| Blood ketones | What to do |
|---|---|
| Below 0.6 mmol/L | Normal. Correct glucose as usual. |
| 0.6 to 1.5 mmol/L | Take a correction dose of Humalog, drink 250 to 500 mL of sugar-free fluid per hour, recheck in 2 hours. |
| 1.5 to 3.0 mmol/L | Call our clinic or on-call endocrinologist now; you are at high risk of DKA. |
| Above 3.0 mmol/L, or vomiting and unable to keep fluids down | Go to the emergency department. |
Rule 4: carbohydrate still matters. If you cannot eat, sip fluids containing some sugar so you can keep taking insulin safely.
Follow-up
- Endocrine clinic with me in 1 to 2 weeks, with a repeat basic metabolic panel. We will also talk about a continuous glucose monitor; many plans cover them, and trend alerts catch problems early.
- The pharmacist has confirmed your Lantus is enrolled in the manufacturer program at 35 dollars a month, and your first box is filled.
If insulin ever becomes hard to afford again, call us before you change a dose. That phone call is far easier than tonight was.
]
"common_mistakes": [
{"mistake":"Stopping or sharply reducing the insulin infusion once glucose normalizes."}— Glucose falls faster than ketones clear. When glucose drops below about 250 mg/dL, dextrose should be added and the infusion reduced (for example to 0.05 units/kg/h) rather than stopped, continuing until beta-hydroxybutyrate is below 0.6 mmol/L and pH is at least 7.3 or bicarbonate at least 18 mmol/L. Stopping early causes rebound ketoacidosis.
(4 more items elided here; the record has 5)
]
"differential_diagnosis": [
{"condition":"Hyperosmolar hyperglycemic state (HHS)"}— Glucose usually above 600 mg/dL with effective osmolality above 300 mOsm/kg, minimal ketosis (beta-hydroxybutyrate below 3.0 mmol/L), pH typically 7.3 or higher and bicarbonate 15 mmol/L or more; more common in older adults with type 2 diabetes and develops over days. Mixed DKA-HHS presentations occur.
(4 more items elided here; the record has 5)
]
"related_diseases": [
{"disease":"Type 1 diabetes mellitus"}— The underlying condition in most DKA; absolute insulin deficiency from autoimmune beta-cell loss means any interruption of insulin, including cost-related rationing, can precipitate ketoacidosis within a day or two.
(4 more items elided here; the record has 5)
]
"executive_summary": Diabetic ketoacidosis (DKA) is an insulin-deficiency emergency defined by hyperglycemia or known diabetes, beta-hydroxybutyrate at or above 3.0 mmol/L, and pH below 7.3 or bicarbonate below 18 mmol/L. This case follows a 26-year-old woman with type 1 diabetes and severe DKA (pH 7.20, bicarbonate 8 mmol/L, anion gap 27) precipitated by cost-driven rationing of basal insulin after losing parental insurance, an omission she initially concealed and that was identified from her pen's open date and remaining volume.
Management principles: 0.9% saline resuscitation (1 L in the first hour, then 250 mL/h for a 58 kg patient); IV regular insulin at 0.1 units/kg/h, reduced to 0.05 units/kg/h with added dextrose once glucose falls below 250 mg/dL; potassium replacement once below 5.0 mmol/L and insulin held if below 3.5 mmol/L; bicarbonate only for pH below 7.0; resolution defined by beta-hydroxybutyrate below 0.6 mmol/L with pH at least 7.3 or bicarbonate at least 18 mmol/L; and subcutaneous glargine given 1 to 2 hours before stopping the infusion.
Key teaching points: apparent hyperkalemia masks a total-body potassium deficit; saline-induced hyperchloremic acidosis can keep bicarbonate low after ketones clear; SGLT2 inhibitors can cause euglycemic DKA, especially in type 1 diabetes; thiazides raise insulin requirements; and preventing recurrence requires addressing insulin affordability and teaching sick-day ketone monitoring.
}
Sample record from the Opus 4.8 set (for comparison)
The sample record from the Opus 4.8 card, copied from that card and rendered the same way, list fields cut to their first item here too — a CDC ring-vaccination response to a confirmed deliberate orthopoxvirus release, with Dr. Arjun Venkatesan counseling the vaccine-hesitant Caldwell family of Pawhuska, Oklahoma. Note the shorter encounter (24 turns) and the old verbatim pubmed_refs shape with summary and no resolved PMIDs.
{
"name": "Smallpox"
"icd10": "B03"
"body_systems": ["integumentary","immune"]
(verbatim source fields omitted here for brevity: aliases, search, description, related_drugs, drug_interactions, food_interactions, prevalence, source_disease)
"pubmed_refs": [
{"pmid":"24507613","year":2014}PMID 24507613 — Smallpox: history, pathogenesis, and contemporary relevance — Review of WHO eradication program and current biodefense preparedness including stockpiled vaccines.{"pmid":"29742189","year":2018}PMID 29742189 — Tecovirimat for orthopoxvirus infections — Phase 3 approval trial data for tecovirimat showing safety and antiviral activity in animal models.]
"clinician_persona": Dr. Arjun Venkatesan is an Indian-American infectious disease physician and the county public health vaccination officer leading a CDC-coordinated ring-vaccination response at a temporary clinic in rural Osage County, Oklahoma, following a confirmed deliberate orthopoxvirus release.
"patient_scenario": The Caldwell family of Pawhuska, Oklahoma: Cody Caldwell, an 8-year-old white boy (the identified contact), accompanied by his parents Wyatt Caldwell, 41, a propane delivery driver, and Brittany Caldwell, 38, a homeschool parent. All three were at a county fairground now linked to a confirmed smallpox exposure event and have been asked to report for post-exposure assessment and vaccination. The parents are deeply skeptical of vaccines and of the clinic, and repeatedly cite Robert F. Kennedy Jr.
"conversation": [ (24 turns — patient turns quoted, clinician turns plain)
{"role":"system"}
You are an Indian-American infectious disease physician and county vaccination officer running a post-exposure ring-vaccination clinic in rural Oklahoma after a confirmed deliberate smallpox (variola) release. Educate a vaccine-hesitant white family who distrust you and quote anti-vaccine media. Be patient, factual, non-condescending, and clinically precise. Use real vitals, labs with reference ranges, exact drugs with dosing, and cite PubMed IDs. No emojis.
{"role":"user"}
Look, we only came because the sheriff's office said we had to register. I'm Wyatt Caldwell, this is my wife Brittany and our boy Cody. I'll be straight with you, doc, we don't do vaccines in this house. We've read the stuff. Kennedy's been saying for years these things are full of toxins and they don't even test them. So you can check the boy over, but we're not getting jabbed with anything.
{"role":"assistant"}
Thank you for coming in, Mr. Caldwell, and I appreciate you being straight with me. I am Dr. Arjun Venkatesan, the county's vaccination officer. Nobody is getting anything against their will today. My job is to give you accurate information and then let you make the decision. Let me start with why you were called in, because it is not routine. Three days ago the FBI and CDC confirmed that someone deliberately released variola virus, the virus that causes smallpox, at the fairground your family attended on Saturday. This is not the seasonal flu shot conversation. Smallpox was declared eradicated in 1980, and the only reason it is back is that it was released on purpose as a weapon (PMID: 24507613). Before we talk vaccines at all, I want to examine Cody and take some vitals. Is that alright?
{"role":"user"}
Wait. Real smallpox? Like the thing from the history books? You're not just saying that to scare us into the shot?
{"role":"assistant"}
I understand the suspicion, and it is a fair question. I am telling you this because it is true and because you have a right to know your family's risk. The lab confirmation came back by PCR and electron microscopy from the index patient at a CDC reference lab; the case is real and the exposure window is real. Here is what matters for you: the virus has an incubation period of about 7 to 17 days, averaging 12. Cody was potentially exposed 3 days ago. That means there is still a window where a vaccine can stop the disease before it starts. Let me check him first. Cody, can you sit up here for me?
{"role":"assistant"}
Cody's vitals: temperature 37.0 C (98.6 F), heart rate 88 bpm, respiratory rate 18/min, BP 102/64 mmHg, SpO2 99% on room air, weight 26 kg. Skin exam: no rash, no oral lesions, mucous membranes clear. He looks completely well, which is exactly what we expect this early. Mr. and Mrs. Caldwell, Cody is currently healthy. The decision in front of you is whether to protect him during the incubation window or to wait and see, and I want to walk you through both honestly.
{"role":"user"}
Here's the thing though. Kennedy says natural immunity is always better than a vaccine. Cody's a strong kid, he's been sick plenty and he bounces right back. Why would we put some lab-made thing in him when his own body can handle it?
{"role":"assistant"}
I hear that argument a lot, and for many ordinary childhood illnesses there is a kernel of truth that recovering from a mild infection gives durable immunity. The problem is that smallpox is not a mild childhood illness. To get the 'natural immunity' you are describing, Cody would first have to survive smallpox itself. Historically, variola major killed about 3 of every 10 people who caught it, and in unvaccinated young children the death rate was higher. Survivors were often left with deep facial scarring and sometimes blindness. So the choice is not 'vaccine versus natural immunity.' It is 'vaccine versus risking a disease that, before vaccination existed, killed an estimated 300 million people in the 20th century alone.' Natural immunity here has a body count attached to it.
{"role":"user"}
Brittany here. I'll be honest, I don't love that you're the one telling us this. No offense, but these agencies and the drug companies, they have an agenda. How do we know you're not just pushing whatever they told you to push?
{"role":"assistant"}
No offense taken, Mrs. Caldwell, and I would rather you ask me that to my face than sit on it. Here is my honest answer. I do not get paid more if your family is vaccinated; I am a salaried public health physician and I would tell you the exact same thing if I were treating my own niece, who is about Cody's age. I am also not going to oversell this to you. The vaccine I would recommend has real risks, and I am going to tell you every one of them before you decide, including the rare bad ones. A salesman hides the downsides. I am going to hand them to you. Does that sound fair?
{"role":"user"}
...Alright. Tell us the downsides then. All of them. Because Kennedy says the side effects are worse than the disease and they cover them up.
{"role":"assistant"}
Good, this is the right conversation to have. There are two different vaccines, and the difference matters enormously. The old one, ACAM2000, contains a live, replicating virus called vaccinia. It is given by scratching the skin with a bifurcated needle, and it works, but it has real risks: myocarditis or pericarditis (heart inflammation) in roughly 1 in 175 first-time recipients in military data (PMID: 14532315), plus a danger called eczema vaccinatum in people with eczema, and the live virus can even spread from the vaccination site to other people in the household. The newer one, JYNNEOS, uses a non-replicating virus (modified vaccinia Ankara). It cannot replicate in the body, so it does not carry that myocarditis signal, it is safe in people with eczema, and it cannot spread to others. It is given as two 0.5 mL injections under the skin, 28 days apart. For a child and for a family like yours, JYNNEOS is what I would recommend precisely because it is the safer of the two. So when you say the side effects can be serious, you are right about the old vaccine, and that is exactly why I am steering you to the safer one rather than away from protection altogether.
{"role":"user"}
Okay, that's... actually more honest than I expected. But the boy's never had a vaccine, so what about us? You said it could spread from the site. We've all been around the same fairground.
{"role":"assistant"}
Correct, and this is the key point. Because all three of you were potential contacts, the recommendation is to vaccinate all three of you, which is why I want to do a quick health check on you and Mr. Caldwell too. The spreading risk I mentioned is specific to the old ACAM2000 vaccine. With JYNNEOS, there is no replicating virus, so there is nothing to spread, which is another reason it is the better fit for a whole household at once. Mr. Caldwell, before I draw anything up, I need to ask you both some standard medical questions. Any heart problems, any history of chest pain, palpitations, or heart medication?
{"role":"user"}
Nope. Healthy as a horse, never been to a doctor in years. Never had a heart issue in my life. Let's just get on with it.
{"role":"assistant"}
Let me get a baseline set on you while we talk. Mr. Caldwell's vitals: BP 146/92 mmHg, heart rate 78 bpm but I am noting a few irregular beats, weight 104 kg. While I had the cuff on you I felt an occasional skipped beat. I am not trying to catch you out, but I want to ask once more, gently: no palpitations, no medication at all? Even something you picked up at the pharmacy or that a doctor started a while back?
{"role":"user"}
...Alright, fine. I take metoprolol. A doctor put me on it a couple years back after my heart was racing and skipping, said it was an arrhythmia. I didn't mention it because I figured you'd use it as an excuse to push more pills on me. But I didn't think it mattered for a smallpox shot.
{"role":"assistant"}
Thank you for telling me, and I want you to notice something: that information makes me more cautious, not less, which is the opposite of what 'they just want to jab everyone' would predict. Your metoprolol and your underlying arrhythmia are exactly why I would never give you the old ACAM2000 vaccine, because its myocarditis risk would be a genuine concern in a man with a known heart rhythm problem and a BP of 146/92. The JYNNEOS vaccine does not carry that cardiac signal, so it remains a reasonable option for you, but I will document your cardiac history, get a baseline ECG today, and check a troponin (normal less than 0.04 ng/mL) so we have a comparison if you ever feel chest symptoms. Concealing a heart condition to avoid pills is the kind of thing that can actually get someone hurt; transparency is what lets me protect you properly.
{"role":"user"}
Okay. I should've said something. So if Cody or any of us actually got sick with this, what then? Is there even a treatment or is it just a death sentence like the history books?
{"role":"assistant"}
There is treatment now, which there was not in the eradication era. The first-line antiviral is tecovirimat, brand name TPOXX, an oral drug that blocks a viral envelope protein and was approved based on the animal-rule efficacy and human safety data (PMID: 29742189). For an adult 40 to 120 kg the dose is 600 mg by mouth twice daily for 14 days; for a child Cody's size, around 26 kg, the weight-based dose would be 400 mg twice daily. One important caution: tecovirimat induces the liver enzyme CYP3A4, so it can lower the levels of other drugs. For example, if Cody ever needed sedation with midazolam for a procedure, the midazolam could be less effective and the dose would need adjusting. There are also second-line antivirals, brincidofovir and cidofovir. Cidofovir is hard on the kidneys, and we must never combine it with aminoglycoside antibiotics like gentamicin, because stacking two kidney-toxic drugs can cause acute kidney injury. But the point is, if prevention fails, we are not helpless anymore.
{"role":"user"}
Brittany again. I think... I think we've been getting a lot of our information from people who weren't going to be standing in front of us if our kid actually got sick. If we do the JYNNEOS one, the non-replicating one, for all three of us, what does that actually look like?
{"role":"assistant"}
That is a very fair way to put it, and I respect you saying it. Here is the plan. Today, each of you gets a single 0.5 mL JYNNEOS injection under the skin of the upper arm. Because you may already be in the early exposure window, giving it now, within roughly 4 days of exposure, gives the best chance of preventing smallpox or making it far milder if it does appear. You come back in 28 days for the second 0.5 mL dose to complete the series. Expect a sore arm, maybe some redness and a low-grade fever for a day or two, which is the normal immune response, not the disease. I will give you a sheet with warning signs to watch for in all three of you over the next two to three weeks: a fever above 38.3 C (101 F), severe headache or backache, and especially a rash with firm, deep bumps that all look the same age, which is the pattern that distinguishes smallpox from chickenpox. If any of that appears, you call the number on the sheet immediately and we start tecovirimat. Does that plan feel like something you can both agree to?
{"role":"user"}
Wyatt here. Yeah. Yeah, do it. Start with me so Cody sees it's not a big deal. And... thanks for not talking down to us. Nobody else bothered to actually explain any of this.
{"role":"assistant"}
That means a lot, Mr. Caldwell, and starting with you so Cody sees his dad go first is exactly the right instinct. Let me document the consent for all three of you, note your cardiac history and metoprolol in your chart, and I will draw up the three JYNNEOS doses. You are making a careful, informed decision for your family, and that is all I have ever wanted out of this conversation. Roll up your left sleeve for me.
]
"common_mistakes": [
{"mistake":"Recommending ACAM2000 to a household with cardiac history or eczema without considering JYNNEOS"}— ACAM2000 is a replication-competent vaccinia vaccine carrying risks of myocarditis/pericarditis (~1 in 175 first-time recipients), eczema vaccinatum, and contact transmission to household members. JYNNEOS, a non-replicating MVA vaccine, avoids these and is preferred for patients with cardiac disease, atopic dermatitis, immunocompromise, or whole-household post-exposure vaccination.
(4 more items elided here; the record has 5)
]
"differential_diagnosis": [
{"condition":"Varicella (chickenpox)"}— Lesions appear in successive crops of different ages, are superficial and centripetal (trunk-predominant), and the patient is usually not severely prostrated. Variola lesions are synchronous, deep-seated, firm, umbilicated, and centrifugal with a severe febrile prodrome.
(4 more items elided here; the record has 5)
]
"related_diseases": [
{"disease":"Mpox (monkeypox)"}— A zoonotic orthopoxvirus in the same genus; JYNNEOS and tecovirimat are used for both, and clinical/vaccine experience overlaps substantially.
(4 more items elided here; the record has 5)
]
"executive_summary": A vaccine-hesitant white Oklahoma family presents for post-exposure assessment during a CDC ring-vaccination response after a confirmed deliberate variola (smallpox) release. The parents distrust the Indian-American vaccination officer and cite anti-vaccine media. Through transparent disclosure of real vaccine risks, the clinician distinguishes the replication-competent ACAM2000 vaccine (myocarditis ~1/175, eczema vaccinatum, contact spread) from non-replicating JYNNEOS (0.5 mL SC, two doses 28 days apart), and recommends JYNNEOS for the whole household. A baseline exam catches an irregular pulse and elevated BP, prompting the father's disclosure of concealed metoprolol use for an arrhythmia, which reinforces avoiding ACAM2000. Treatment contingencies include tecovirimat 600 mg PO BID x14 days (400 mg BID weight-based for the 26 kg child) with a CYP3A4-induction caution for midazolam, and second-line brincidofovir/cidofovir with avoidance of concurrent aminoglycosides. Honest, non-condescending counseling converts refusal into informed consent for all three family members.
}
If you see anything wrong please be constructive about it and send in a pull request to fix it. Love you.
Author: Nisten Tahiraj
License: MIT
Contributors
2 commits
Similar repos
Opus-5.5 generated Doctor-Patient Conversations for All Human Diseases
8
2 commits
updated Sep 26, 2026
README
Opus-5.5 generated Doctor-Patient Conversations for All Human Diseases
The sequel to nisten/opus-doctor-patient-conversations-all-human-diseases (Opus 4.8). Same disease list, same 20-key schema, same ChatML conversations — regenerated from scratch with Claude Opus 5.5, one agent per disease, and held to a much stricter bar.
Covers every human disease listed on my previous work here: nisten/all-human-diseases
What changed since the Opus 4.8 set: the conversations are about 60% longer (avg 33 turns, ~5,700 tokens vs 29 turns, ~3,500) and run from presentation through workup, treatment and disposition with real vitals and labs, units and reference ranges included. Every PubMed ID cited in the text was checked live against NCBI before a record was accepted, and pubmed_refs is rebuilt per record from the citations the conversation actually makes (the curated source lists turned out to be mostly wrong: only 5.8% of their PMIDs point at the paper their title claims). Internal consistency was enforced, not just format: blood gases must satisfy Henderson-Hasselbalch, the anion gap must match the chemistry panel, doses must match the stated weight, and drugs must come from the record's own related_drugs.
Each conversation simulates a realistic patient from a randomly chosen diverse ethnic background, and in each one there is exactly one obfuscation or outright lie from the patient which the doctor tactfully catches and corrects. The drug-drug and drug-food interactions were all grounded via Health Canada Open Data: https://health-products.canada.ca/api/documentation/dpd-documentation-en.html
This is NOT a distillation, the agent traces are NOT included, only the final verified records are here, meant to be used as a source of truth and medical safety when training or using medical AI models & apps <3
- Records: 2,194 clinical encounters covering 2,194 unique human diseases
- Format: one JSON object per line (JSONL), 20 keys per record, 75 MB
- Conversations: avg 33.4 turns (25-59) and ~22,800 chars (~5,700 tokens) each; ~12.5M conversation tokens total
- Citations: 6,377 verified PubMed references across the records (4,198 distinct PMIDs), 2-5 per record, every one resolving on NCBI with a matching title
- Language: English
Provenance
- The seed list of disease names (in canonical order) is derived from nisten/all-human-diseases.
- The structured clinical fields (
description,related_drugs,drug_interactions,food_interactions,icd10,prevalence,body_systems) are copied verbatim from the same curated source as the Opus 4.8 set; the model never invents them. Two source texts were corrected after the fact: the methimazole-warfarin interaction description in three thyroid records (it had the INR direction backwards) and one trial summary (EMBARK's primary endpoint was not met). pubmed_refsis rebuilt, not verbatim. The curated source lists carried 3,925 distinct PMIDs; checked one by one against NCBI, 226 point at the paper their title claims, 3,582 resolve to an unrelated paper and 117 do not exist. Rather than ship that, each record'spubmed_refsnow lists the PMIDs the record itself cites, as{pmid, title, year}with the title taken from NCBI. Thesummaryfield of the old shape is gone (it was the fabricated part).- The 7 creative fields (personas, scenario, conversation, common mistakes, differential, related diseases, executive summary) were generated and then verified agentically by Claude Opus 5.5 subagents, one disease at a time. Each record passed automated gates before being committed: all 20 keys present, strict role alternation, >=15 turns, >=8,000-char conversation, >=3 items in every list field, zero emoji, no chatbot filler openers, no leaked generation instructions, and a live NCBI title check on every PMID in the text. Demographics were rolled per record from weighted workforce and population tables so clinician and patient backgrounds differ realistically.
- Two caveats, stated once: of the 2,195 names on the seed list one is excluded because the model's safety classifier stopped every generation attempt on it and I did not route around that, and the last six lines of the file were regenerated on a retry through the harness's generic
opusalias, so they are Opus-family but not verifiably 5.5. - The agentic traces / reasoning are NOT included — only the final, validated records.
Intended use
Built to be a source-of-truth dataset for RAG systems (clinical Q&A, retrieval-augmented assistants) and for training / fine-tuning small models on grounded, structured-yet-conversational medical dialogue. It is not medical advice, you're responsible for your own use of it.
Schema (20 keys)
| Key | Type | Source | Notes |
|---|---|---|---|
_type | string | — | always disease_chatml |
name | string | seed | canonical disease name |
aliases | string[] | source | synonyms |
search | string[] | source | search terms |
description | string | verbatim | clinical description |
related_drugs | object[] | verbatim | {din, brand_name, ingredient, atc_code, role} |
drug_interactions | object[] | verbatim | {drug_a, drug_b, severity, description} |
food_interactions | object[] | verbatim | may be empty for some diseases |
pubmed_refs | object[] | rebuilt | {pmid, title, year} — the PMIDs this record cites, title from NCBI |
icd10 | string | verbatim | ICD-10 code |
prevalence | string | verbatim | epidemiology text |
body_systems | string[] | verbatim | e.g. ["dermatologic"] |
source_disease | string | seed | name as keyed |
clinician_persona | string | generated | named physician, specialty, background |
patient_scenario | string | generated | named patient, presentation |
conversation | object[] | generated | ChatML-ready {role, content}, roles: system/user/assistant |
common_mistakes | object[] | generated | clinical pitfalls |
differential_diagnosis | object[] | generated | ranked differentials |
related_diseases | object[] | generated | adjacent conditions |
executive_summary | string | generated | concise case summary |
Disease coverage (2,194 unique diseases)
All disease names in the dataset, alphabetical, read top-to-bottom then left-to-right (scroll horizontally):
22q11.2 deletion syndrome Blastocystosis Constipation in children Fibrosis, pulmonary Incontinence, fecal Myopia Pressure sore Stings, jellyfish
A fib Blepharitis Contact dermatitis Fifth disease Incontinence, nighttime Myxofibrosarcoma Pressure ulcer Stings, scorpion
Abdominal aortic aneurysm Blocked tear duct Conversion disorder Filoviruses Incontinence, urinary NAFLD Preterm labor Stomach cancer
Abnormally excessive sweating Blocked ureter Cooley's anemia Fistula, arteriovenous Indigestion Nail fungus Priapism Stomach flu
Abscess, Bartholin's Blood in urine (hematuria) COPD Flatfeet Infant jaundice Narcissistic personality disorder Prickly heat Stomach pain, nonulcer
Absence seizure BMS Corns and calluses Flaviviruses Infant reflux Narcolepsy Primary adrenal insufficiency Stomach polyps
Acanthosis nigricans Body dysmorphic disorder Coronary artery disease Floaters Infantile hemangioma Nasal and paranasal tumors Primary biliary cholangitis Stomatodynia
Achalasia Body lice Coronary microvascular disease Floor of the mouth cancer Infectious arthritis Nasal polyps Primary biliary cirrhosis Stool holding
Achilles tendinitis Body odor and sweating Coronavirus disease 2019 (COVID-19) Floppy valve syndrome Infectious diseases Nasopharyngeal carcinoma Primary ciliary dyskinesia Strawberry hemangioma
Achilles tendon rupture Boils and carbuncles Corticobasal degeneration (corticobasal syndrome)Flu Infertility Nearsightedness Primary immunodeficiency Strep throat
Acid reflux Bone cancer Costochondritis Flu, avian Infertility, female Neck pain Primary lateral sclerosis (PLS) Stress cardiomyopathy
Acid reflux, infant Bone infection Costosternal chondrodynia Flu, bird Infertility, male Nephroblastoma Primary ovarian insufficiency Stress fractures
ACL injury Bone metastasis Costosternal syndrome Flu, swine Inflamed gallbladder Nephrogenic systemic fibrosis Primary polycythemia Stress incontinence
Acne Bone spurs Cough headaches Fluid around the heart Inflamed pancreas Nephrotic syndrome Primary progressive aphasia Stretch marks
Acne inversa Borderline personality disorder Cough, chronic FND Inflamed pericardium Neuralgia, postherpetic Primary sclerosing cholangitis Stroke
Acoustic neuroma Botulism COVID-19 Focal segmental glomerulosclerosis (FSGS)Inflammatory bowel disease (IBD) Neuralgia, trigeminal Proctitis Stump pain
Acquired immunodeficiency syndrome Bowel incontinence COVID-19-associated multisystem inflammatory syndrome in childrenFolate deficiency anemia Inflammatory breast cancer Neurilemmoma Progeria Stuttering
Acromegaly Bowel obstruction Crabs Folliculitis Influenza (flu) Neurinomatosis Progressive supranuclear palsy Stye (sty)
Actinic keratosis BPH Cradle cap Food allergy Influenza, avian Neuroblastoma Prolactinoma Subarachnoid hemorrhage
Acute coronary syndrome Brachial plexus injury Cramp, muscle Food allergy, egg Influenza, H1N1 Neurocardiogenic syncope Prolapse, mitral valve Subconjunctival hemorrhage (broken blood vessel in eye)
Acute flaccid myelitis (AFM) Bradycardia Cramps, menstrual Food allergy, milk Influenza, swine flu Neurodermatitis Prolapsed bladder Subfertility
Acute granulocytic leukemia Bradycardia-tachycardia syndrome Cranial arteritis Food poisoning Ingrown hair Neuroendocrine carcinoma of the skin Prolapsed uterus Substance abuse disorder
Acute inflammatory demyelinating polyneuropathyBrain aneurysm Craniopharyngioma Food-borne illness Ingrown toenails Neuroendocrine tumors Prostate cancer Sudden cardiac arrest
Acute kidney injury Brain arteriovenous malformation Craniosynostosis Foot drop Inguinal hernia Neurofibroma Prostate gland enlargement Sudden infant death syndrome (SIDS)
Acute liver failure Brain AVM (arteriovenous malformation) Creutzfeldt-Jakob disease Foot fracture Inherited metabolic disorders Neurofibromatosis type 1 Prostatitis Suicide
Acute lymphoblastic leukemia Brain metastases Crib death Foramen ovale Insomnia Neurofibrosarcoma Pruritis ani Suicide and suicidal thoughts
Acute lymphocytic leukemia Brain tumor Crohn's disease Fracture, arm Insulin resistance syndrome Neuroma, acoustic Pruritus Sun allergy
Acute lymphoid leukemia Brain tumor, child Croup Fracture, growth plate Intermittent claudication Neuroma, Morton's Pseudobulbar affect Sun poisoning
Acute myeloblastic leukemia Breast cancer CRPS Fracture, hip Intermittent explosive disorder Neuroma, plantar Pseudocholinesterase deficiency Sunburn
Acute myelogenous leukemia Breast cancer, inflammatory Cryoglobulinemia Fracture, leg Interstitial cystitis Neuromyelitis optica Pseudogout Sunken chest
Acute myeloid leukemia Breast cancer, male Cryptorchidism Fracture, stress Interstitial lung disease Neuromyelitis optica spectrum disorder Pseudomembranous colitis Supranuclear palsy
Acute nonlymphocytic leukemia Breast cysts CSF leak (Cerebrospinal fluid leak) Fractured nose Intestinal ischemia Neuropathy, autonomic Pseudotumor cerebri (idiopathic intracranial hypertension)Supraventricular tachycardia
Acute radiation sickness Breast infection CTCL Fractured rib Intestinal lipodystrophy Neuropathy, diabetic Psoriasis Suspicious breast lumps
Acute radiation syndrome Breast pain Curvature of the penis Frontal lobe epilepsy Intestinal obstruction Neuropathy, hereditary motor and sensoryPsoriatic arthritis SVT
Acute renal failure Broken ankle Curvature of the spine Frontal lobe seizures Intracranial hematoma Neuropathy, peripheral PTSD Sweating and body odor
Acute respiratory distress syndrome Broken arm Cushing syndrome Frontotemporal dementia Intracranial venous malformations Nevus Pubic lice (crabs) Sweating, abnormally excessive
Acute sinusitis Broken blood vessel in eye Cutaneous B-cell lymphoma Frontotemporal lobar degeneration Intraductal carcinoma Nickel allergy Puffy eyes Swelling
Addiction, alcohol Broken collarbone Cutaneous T-cell lymphoma Frostbite Intrahepatic cholestasis of pregnancy Nicotine addiction Pulmonary atresia Swimmer's ear
Addiction, gambling Broken foot Cutting/self-injury Frozen shoulder Intussusception Nicotine dependence Pulmonary atresia with intact ventricular septumSwimmer's itch
Addiction, nicotine Broken hand CVA FSGS Invasive lobular carcinoma Niemann-Pick disease Pulmonary atresia with ventricular septal defectSwollen knee
Addison's disease Broken heart syndrome CVID Fuchs dystrophy Iritis Nightmare disorder Pulmonary edema Swollen lymph nodes
Adenitis, mesenteric Broken hip Cyclic vomiting syndrome Functional dyspepsia Iron deficiency anemia Nighttime incontinence Pulmonary embolism Syndrome X
Adenomyosis Broken leg Cyclothymia (cyclothymic disorder) Functional neurologic disorder/conversion disorderIron overload NMO Pulmonary fibrosis Synostosis
ADHD in children Broken nose Cyclothymic disorder Functional neurological symptom disorderIrritable bowel syndrome NMOSD Pulmonary hypertension Synovial sarcoma
ADHD, Adult Broken ribs Cyst, Bartholin's Fungal infection, nail Ischemia, intestinal Noctural enuresis Pulmonary valve disease Syphilis
Adhesive capsulitis Broken toe Cyst, ganglion Funnel chest Ischemic bone necrosis Noise-related hearing loss Pulmonary valve stenosis Syringomyelia
Adjustment disorders Broken wrist Cyst, kidney GAD Ischemic colitis Non-Hodgkin's lymphoma PVCs Systemic exertional intolerance disease
Adnexal tumors Bronchiolitis Cyst, ovarian Galactorrhea Ischemic priapism Nonalcoholic fatty liver disease PVD Systemic lupus erythematosus
Adolescent schizophrenia Bronchitis Cyst, pancreatic Gallbladder cancer Islet cell cancer Nonallergic rhinitis PVPS Systemic mastocytosis
Adrenal cancer Brucellosis Cyst, pilonidal Gallbladder inflamation Itchy skin (pruritus) Nonischemic priapism Pyelonephritis Systemic sclerosis
Adrenal mass Brugada syndrome Cyst, spermatic Gallstones ITP Nonmelanoma skin cancer Pyloric stenosis Tachycardia
Adrenoleukodystrophy Bruxism (teeth grinding) Cystic fibrosis Gambling, compulsive Jaw tumors and cysts Nontropical sprue Pyoderma gangrenosum Takayasu's arteritis
Adult attention-deficit/hyperactivity disorder (ADHD)Buerger disease Cystitis Ganglion cyst Jellyfish stings Nonulcer dyspepsia RA Takotsubo cardiomyopathy
Adult Still disease Bulimia nervosa Cystitis, interstitial Gangrene Jet lag disorder Noonan syndrome Rabies Tapeworm infection
AFM Bullous pemphigoid Cystocele Gas and gas pains Jock itch Norovirus infection Radiation enteritis TAPVR
Age spots (liver spots) Bundle branch block Cysts, epidermoid Gastric cancer Jumper's knee Nose fracture Radiation sickness Tay-Sachs disease
Age-related macular degeneration, dry Bunions Cysts, sebaceous Gastric emptying, delayed Juvenile idiopathic arthritis Nymphomania Ramsay Hunt syndrome TB
Age-related macular degeneration, wet Bunyaviruses Cytomegalovirus (CMV) infection Gastric emptying, rapid Juvenile rheumatoid arthritis OAB Rapid gastric emptying TBI
Agnogenic myeloid metaplasia Burn injury Cytomegalovirus infection Gastritis Juvenile schizophrenia Obesity Rapid heartbeat Tears, decreased production
Agoraphobia Burning mouth syndrome Dandruff Gastroenteritis, viral Kaposi sarcoma Obesity, childhood Rash, poison ivy Teen depression
AIDP Burning thigh pain Daytime sleepiness Gastroesophageal reflux disease (GERD) Kawasaki disease Obsessive-compulsive disorder (OCD) Raynaud's disease Teeth grinding
AIDS/HIV Burns DCIS Gastrointestinal bleeding Keloid scar Obstetric cholestasis Reactive arthritis Temporal arteritis
Airplane ear Bursitis DDH Gastrointestinal stromal tumor (GIST) Keratitis Obstructive sleep apnea Reactive attachment disorder Temporal lobe seizure
Albinism Bursitis of the knee De Quervain tenosynovitis Gastroparesis Keratoconjunctivitis sicca Obstructive sleep apnea in children Reading disability, specific Temporomandibular disorders
Alcohol addiction C. difficile colitis Decreased tear production Gaucher disease Keratoconus Occupational asthma Rebound headache Temporomandibular joint disorders
Alcohol intolerance C. difficile infection Deep vein thrombosis (DVT) GBM Keratosis pilaris OCD Rectal cancer TEN
Alcohol poisoning Calciphylaxis Deficient color vision Gender dysphoria Keratosis, actinic Ocular albinism Rectal inflammation Tendinitis
Alcohol use disorder Calluses and corns Degenerative arthritis Gender identity disorder Keratosis, seborrheic Ocular rosacea Rectal itching Tendinitis, Achilles
Alcohol-associated hepatitis Cancer Dehydration Generalized anxiety disorder Ketoacidosis, diabetic Oculocutaneous albinism Rectal prolapse Tendinitis, patellar
Alcoholic hepatitis Cancer, acute lymphocytic leukemia Delayed ejaculation Genital herpes Kidney cancer Oculosympathetic palsy Rectal ulcer Tennis elbow
Allergic granulomatosis Cancer, anal Delayed gastric emptying Genital warts Kidney cysts ODD Rectovaginal fistula Tenosynovitis, de Quervain's
Allergic granulomatosis and angiitis Cancer, basal cell Delayed sleep phase Geographic tongue Kidney disease, chronic Odontogenic tumors and cysts Recurrent breast cancer Tension headache
Allergic rhinitis Cancer, bladder Delayed sleep phase syndrome GERD Kidney failure, acute OHSS Recurrent oral ulcers Testicle, retractile
Allergies Cancer, bone Delayed sleep-wake phase sleep disorder Germ cell tumors Kidney failure, chronic Olfactory neuroblastoma Red eye Testicle, undescended
Allergy, dust mite Cancer, breast Delirium German measles Kidney infection Oligodendroglioma Reflux, bile Testicular cancer
Allergy, egg Cancer, carcinoid tumors Dementia Gestational diabetes Kidney stones Onychomycosis Reflux, vesicoureteral Testicular torsion
Allergy, food Cancer, cervical Dementia, frontotemporal Giant cell arteritis Kissing disease Oppositional defiant disorder (ODD) Regurgitation, aortic valve Testosterone deficiency
Allergy, latex Cancer, chronic lymphocytic leukemia Dementia, Lewy body Giant lymph node hyperplasia Klatskin tumor Optic neuritis Regurgitation, mitral valve Tetanus
Allergy, milk Cancer, chronic myelogenous leukemia Dementia, vascular Giardia infection (giardiasis) Kleptomania Oral cancer Reiter's syndrome Tetralogy of Fallot
Allergy, mold Cancer, colon Dengue fever Gilbert syndrome Klinefelter syndrome Oral candidiasis REM sleep behavior disorder TGA
Allergy, nickel Cancer, endometrial Dengue hemorrhagic fever Gingivitis Klippel-Trenaunay syndrome Oral lichen planus Renal artery stenosis Thalassemia
Allergy, peanut Cancer, esophageal Dependence, nicotine Glaucoma Knee bursitis Oral thrush Renal cancer Thoracic aortic aneurysm
Allergy, penicillin Cancer, eye melanoma Depersonalization-derealization disorderGlioblastoma Knee pain Orchitis Renal failure, acute Thoracic outlet syndrome
Allergy, pet Cancer, gallbladder Depression (major depressive disorder) Glioblastoma multiforme KP Orthostatic hypotension (postural hypotension)Renal lithiasis Three-day measles
Allergy, shellfish Cancer, gastric Depression, postpartum Glioma Kyphosis Osteoarthritis Residual limb pain Throat cancer
Allergy, wheat Cancer, hairy cell leukemia Depression, teen Glomerulonephritis Lactase deficiency Osteoarthritis, cervical Respiratory syncytial virus (RSV) Thromboangiitis obliterans
Alopecia Cancer, Hodgkin's disease Dermatitis Glossodynia Lactation mastitis Osteochondritis dissecans Restless legs syndrome Thrombocytopenia
Alpha-gal syndrome Cancer, Hurthle cell Dermatitis, atopic Glucocerebrosidase deficiency Lactose intolerance Osteogenic sarcoma Retinal detachment Thrombocytosis
ALS Cancer, inflammatory breast Dermatitis, cercarial Gluten-sensitive enteropathy Laryngitis Osteomyelitis Retinal diseases Thrombophlebitis
Alveolar osteitis Cancer, kidney Dermatitis, contact Goiter Lassa fever Osteonecrosis Retinoblastoma Thrush, oral
Alzheimer's disease Cancer, leukemia Dermatitis, scratch Golfer's elbow Lateral elbow tendinopathy Osteophytes Retinopathy, diabetic Thumb arthritis
Ambiguous genitalia Cancer, lip Dermatitis, seborrheic Gonorrhea Lateral epicondylitis Osteoporosis Retractile testicle Thunderclap headaches
Amblyopia Cancer, liver Dermatofibrosarcoma protuberans Gout Lateral epicondylosis Osteosarcoma Retrograde ejaculation Thyroid cancer
Ameloblastoma Cancer, lung Dermatographia (Dermatographism) Granuloma annulare Latex allergy Otitis externa Rett syndrome Thyroid gland enlargement
Amenorrhea Cancer, male breast Dermatographism Granulomatosis with polyangiitis Lazy eye (amblyopia) Otitis media Reye's syndrome Thyroid nodules
American trypanosomiasis Cancer, Merkel cell Dermatomyositis Graves' disease LCIS Outer ear infection Rhabdomyosarcoma Thyroid, overactive
Amnesia Cancer, mesothelioma Dermatophytosis Grief Lead poisoning Ovarian cancer Rheumatic fever Thyroiditis, chronic lymphocytic
Amnesia, transient global Cancer, mouth Desmoid tumors Gross hematuria Left ventricular hypertrophy Ovarian cysts Rheumatoid arthritis TIA
Amnestic syndrome Cancer, multiple myeloma Desmoplastic small round cell tumors Group B strep disease Leg fracture Ovarian hyperstimulation syndrome Rhinitis, allergic Tic douloureux
Ampullary cancer Cancer, nasopharyngeal Detached retina Growing pains Legg-Calve-Perthes disease Overactive bladder Rhinitis, nonallergic Tinea barbae
Amyloid disease Cancer, neuroblastoma Developmental dysplasia of the hip Growth hormone excess Legionellosis Overactive thyroid Rhinitis, vasomotor Tinea capitis
Amyloidosis Cancer, non-Hodgkin's lymphoma Deviated septum Growth plate fractures Legionnaires' disease Oxyphil cell carcinoma Rib fracture Tinea corporis
Amyotrophic lateral sclerosis (ALS) Cancer, oral Devic's disease Guillain-Barre syndrome Leiomyosarcoma Pa Rickets Tinea cruris
Anal cancer Cancer, ovarian DFSP Gum disease, gingivitis Lentigines, solar PAD Rift Valley fever Tinea pedis
Anal fissure Cancer, pancreatic Diabetes Gum disease, periodontitis Leukemia PAES Ringing in the ear Tinea versicolor
Anal fistula Cancer, paraneoplastic syndromes Diabetes insipidus Gynecomastia Leukemia, acute lymphocytic Paget's disease of bone Ringworm (body) Tinnitus
Anal itching Cancer, prostate Diabetes mellitus H. pylori infection Leukemia, acute myelogenous Paget's disease of the breast Ringworm (scalp) TMD
Anaphylaxis Cancer, rectal Diabetes, gestational H1N1 flu Leukemia, chronic lymphocytic Pain, back Ringworm of the foot TMJ disorders
Anemia Cancer, retinoblastoma Diabetes, type 1 H1N1v flu Leukemia, chronic myelogenous Pain, chest RLS Toe walking in children
Anemia, aplastic Cancer, skin Diabetes, type 1 in children H3N2v flu Leukemia, general Pain, wrist Road rage Toenail fungus
Anemia, Cooley's Cancer, soft tissue sarcoma Diabetes, type 2 Hair loss Leukemia, hairy cell Painful bladder syndrome Rosacea Tongue cancer
Anemia, iron deficiency Cancer, squamous cell Diabetes, type 2 in children Hair-pulling disorder Leukoplakia Painful intercourse (dyspareunia) Roseola Tongue tie
Anemia, Mediterranean Cancer, stomach Diabetic coma Hairy cell leukemia Lewy body dementia Painful periods Rotator cuff injury Tongue-tie (ankyloglossia)
Anemia, sickle cell Cancer, testicular Diabetic hypoglycemia Hairy tongue Lice Palsy, Bell's Rotavirus Tonic-clonic (grand mal) seizure
Anemia, vitamin deficiency Cancer, throat Diabetic ketoacidosis Halitosis Lice, body Palsy, cerebral Round back Tonic-clonic seizure
Aneurysm, abdominal aortic Cancer, thyroid Diabetic nephropathy (kidney disease) Hammertoe and mallet toe Lichen planus Palsy, facial RSV Tonsil cancer
Aneurysm, aortic Cancer, uterine Diabetic neuropathy Hamstring injury Lichen planus, oral Pancreas inflammation Rubella Tonsillitis
Aneurysm, brain Cancer, vagina Diabetic retinopathy Hand fracture Lichen sclerosus Pancreatic cancer Rubeola Tooth abscess
Aneurysm, cerebral Cancer, vulvar Diaper rash Hand-foot-and-mouth disease Lichen simplex chronicus Pancreatic cysts Rumination syndrome Tooth decay
Aneurysm, popliteal Cancer, Wilms' tumor Diarrhea Hangovers Lip cancer Pancreatic neuroendocrine tumors Runner's knee Torn meniscus
Aneurysm, thoracic aortic Candidiasis, oral Diarrhea, antibiotic-associated Hantavirus pulmonary syndrome Lipodystrophy, intestinal Pancreatitis Ruptured disk Torsion, testicular
Aneurysms Candidiasis, vaginal Diarrhea, traveler's Hardening of the arteries Lipoma Panic attacks and panic disorder Ruptured eardrum (perforated eardrum) Torticollis, spasmodic
Angelman syndrome Canker sore Difficulty speaking Hashimoto's disease Liposarcoma PAPVR Ruptured spleen Total anomalous pulmonary venous connection
Angiitis Carbon monoxide poisoning Difficulty swallowing HAV infection Listeria infection Paraganglioma Sacral dimple Total anomalous pulmonary venous return (TAPVR)
Angina Carbuncles and boils DiGeorge syndrome (22q11.2 deletion syndrome)Hay fever Listeriosis Paraneoplastic neurological disorder Sacroiliitis Tourette syndrome
Angina pectoris Carcinoid syndrome Dilated cardiomyopathy HBV infection Liver cancer Paraneoplastic neurological syndrome SAD Toxemia
Angioedema and hives Carcinoid tumors Diphtheria HCC Liver disease Paraneoplastic syndromes of the nervous systemSalivary gland tumors Toxic epidermal necrolysis
Angiofollicular lymph node hyperplasia Carcinoma of unknown primary Dislocated shoulder HCV infection Liver failure, acute Parkinson's disease Salmonella infection Toxic hepatitis
Angiosarcoma Cardiac arrest, sudden Dissecting aneurysm Head and neck cancers Liver hemangioma Parotid tumors Salmonellosis Toxic shock syndrome
Ankle fracture Cardiac ischemia Dissociative disorders Head lice Liver spots Partial anomalous pulmonary venous connectionSarcoidosis Toxoplasmosis
Ankle sprain Cardiogenic shock Diverticulitis Head trauma, coma Liver, enlarged Partial anomalous pulmonary venous returnSarcoma Trachoma
Ankyloglossia Cardiomegaly Dizziness Headache, chronic daily Lobular carcinoma in situ (LCIS) Parvovirus infection Sarcoma, bone cancer Transient global amnesia
Ankylosing spondylitis Cardiomyopathy Double uterus Headache, cluster Localized scleroderma Patellar tendinitis Sarcoma, soft tissue Transient ischemic attack (TIA)
Anorexia Cardiomyopathy, dilated Double-outlet right ventricle Headache, exercise Lockjaw Patellofemoral pain syndrome SBBO Transposition of great vessels
Anorexia nervosa Cardiomyopathy, hypertrophic Down syndrome Headache, migraine Long QT syndrome Patent ductus arteriosus (PDA) Scabies Transposition of the great arteries
Anorgasmia in women Cardiospasm Dressler syndrome Headache, primary cough Loss of bladder control Patent foramen ovale SCAD Transverse myelitis
Anterior cruciate ligament injury Cardiovascular disease Drop foot Headache, sex Lou Gehrig's disease Pathological gambling Scalded mouth syndrome Traumatic brain injury
Anterior vaginal prolapse Carotid artery disease Drug addiction (substance use disorder) Headache, sinus Low blood pressure (hypotension) Pathological laughter and crying Scarlatina Traumatic grief
Anterior vaginal prolapse (cystocele) Carotid artery stenosis Drug allergy Headache, spinal Low blood sugar Pathological stealing Scarlet fever Traveler's diarrhea
Anthrax Carotid insufficiency Dry eyes Headache, tension Low body temperature PBA Schizoaffective disorder Trichinosis
Antibiotic-associated colitis Carpal tunnel syndrome Dry macular degeneration Headache, thunderclap Low platelet count PCOS Schizoid personality disorder Trichomoniasis
Antibiotic-associated diarrhea CAS Dry mouth Headaches in children Low sex drive in women PDA Schizophrenia Trichotillomania (hair-pulling disorder)
Antiphospholipid syndrome Castleman disease Dry skin Health anxiety Low sperm count Peanut allergy Schizophrenia, adolescent, childhood, juvenileTricuspid atresia
Antisocial personality disorder Cataracts Dry socket Hearing loss Low testosterone Pectus excavatum Schizotypal personality disorder Tricuspid valve disease
Anxiety disorder, generalized Cavernous malformations DSPS Heart arrhythmia Low-flow priapism Pediatric brain tumors Schwannoma Tricuspid valve regurgitation
Anxiety disorder, social Cavities and tooth decay DSRCT Heart attack Lung cancer Pediatric obstructive sleep apnea Schwannomatosis Trigeminal neuralgia
Anxiety disorders CDH DSWPD Heart defects in adults Lung collapse Pediatric white blood cell disorders SCI Trigger finger
Aortic aneurysm Celiac artery compression Ductal carcinoma in situ (DCIS) Heart disease Lung disease, interstitial Pediculosis capitis Sciatica Triple X syndrome
Aortic aneurysm, abdominal Celiac artery compression syndrome Dumping syndrome Heart failure Lupus Pelvic inflammatory disease (PID) Scleroderma Trisomy 21 syndrome
Aortic coarctation Celiac axis syndrome Dunbar syndrome Heart murmurs Lupus nephritis Pelvic organ prolapse Sclerosing cholangitis Trisomy X
Aortic dissection Celiac disease Dupuytren contracture Heart palpitations Lyme disease Pelvic pain, chronic Sclerosing mesenteritis Trouble swallowing
Aortic valve disease Celiac trunk compression syndrome Dural arteriovenous fistulas Heart valve disease Lymph nodes, swollen Pelvic support problems, uterine prolapseSclerosis, multiple Truncus arteriosus
Aortic valve regurgitation CELLO Dust mite allergy Heartburn Lymphadenitis Pemphigus Sclerosis, systemic Trypanosomiasis, American
Aortic valve stenosis Cellulite DVT Heat exhaustion Lymphadenitis, mesenteric Penicillin allergy Scoliosis Tuberculosis
APD Cellulitis Dwarfism Heat rash Lymphedema Peptic ulcer Scorpion sting Tuberous sclerosis
Aphasia Central auditory processing disorder Dysarthria Heatstroke Lymphoma Perforated eardrum Scrotal masses Tumor, salivary gland
Aphasia, primary progressive Central nervous system vascular malformationsDyshidrosis Heavy menstrual bleeding Lymphoma, Hodgkin's Pericardial effusion Seasonal affective disorder (SAD) Turner syndrome
Aphthous stomatitis Central sleep apnea Dyslexia Heel pain Lymphoma, non-Hodgkin's Pericardial inflammation Seasonal allergy Tympanic membrane perforation
Aphthous ulcer Cercarial dermatitis Dysmenorrhea Helicobacter pylori (H. pylori) infectionLymphoplasmacytic lymphoma Pericarditis Sebaceous carcinoma Type 1 diabetes
Apical ballooning syndrome Cerebral aneurysm Dyspareunia Hemangioma Lynch syndrome Perimenopause Sebaceous cysts Type 1 diabetes in children
Aplastic anemia Cerebral palsy Dyspepsia Hemangioma, liver Macular degeneration, dry Periodontal disease, gingivitis Seborrheic dermatitis Type 2 diabetes
Appendicitis Cerebrospinal fluid leak Dysphagia Hematoma, intracranial Macular degeneration, wet Periodontal disease, periodontitis Seborrheic keratosis Type 2 diabetes in children
APS Cerebrovascular accident Dysphonia Hematuria Mad cow disease Periodontitis Secondary brain cancer Typhoid fever
ARDS Cerumen impaction Dysrhythmias Hemifacial spasm Major depressive disorder Peripheral artery disease (PAD) Secondary hypertension Ulcer, aphthous
Arenaviruses Cervical cancer Dysthymia Hemochromatosis Malaria Peripheral nerve injuries SEID Ulcer, peptic
Argentine hemorrhagic fever Cervical dystonia Dystonia Hemolytic uremic syndrome (HUS) Male breast cancer Peripheral nerve tumors Seizure disorder Ulcer, rectal
Arm fracture Cervical osteoarthritis Dystonia, cervical Hemophilia Male hypogonadism Peripheral neuropathy Seizure, absence Ulcerative colitis
Arnold-Chiari malformation Cervical pain E. coli Hemorrhoids Male infertility Peripheral vascular disease Seizure, febrile Ulnar wrist pain
Arrhythmia Cervical spondylosis Ear infection (middle ear) Henoch-Schonlein purpura Malignant fibrous histiocytoma Peritonitis Seizure, grand mal Ulnar-sided wrist pain
Arteriosclerosis / atherosclerosis Cervicitis Ear infection, outer ear Hepatitis A virus infection Malignant hyperthermia Pernicious anemia Seizure, petit mal Umbilical hernia
Arteriovenous fistula Chagas disease Eardrum, ruptured Hepatitis B virus infection Malignant mesothelioma Pernio Seizure, temporal lobe Uncontrollable laughter and crying
Arteriovenous malformation Charcot-Marie-Tooth disease Early puberty Hepatitis C virus infection Malignant peripheral nerve sheath tumorsPeroneal muscular atrophy Seizures Under-eye puffiness
Arteritis, giant cell CHD Earwax blockage Hepatitis, autoimmune Mallet toe and hammertoe Persistent depressive disorder Selective IgA deficiency Underactive thyroid
Arteritis, Takayasu's Chemical dependency Eating disorders Hepatitis, toxic MALS Persistent post-concussive symptoms (Post-concussion syndrome)Self harming Undescended testicle
Arthritis Chemo brain Eating disorders, anorexia Hepatocellular carcinoma Mammary duct ectasia Personality disorder, antisocial Self-injury/cutting Undifferentiated pleomorphic sarcoma
Arthritis, basal joint Chest pain Eating disorders, binge eating Hepatolenticular degeneration Manic-depressive illness Personality disorder, borderline Separation anxiety disorder Upset stomach
Arthritis, degenerative Chest wall pain Eating disorders, bulimia Hepatomegaly Marfan syndrome Personality disorder, narcissistic Sepsis Ureteral cancer
Arthritis, gouty Chiari malformation Ebstein anomaly Hepatopulmonary syndrome MASLD Personality disorder, schizoid Septal defect, atrial Ureteral obstruction
Arthritis, infectious Chickenpox Ectopic heartbeat Hereditary hemochromatosis Mastalgia Personality disorder, schizotypal Septal defect, atrioventricular Urethral stricture
Arthritis, juvenile idiopathic Chilblains Ectopic pregnancy Hereditary hemorrhagic telangiectasia Mastitis Personality disorders Septal defect, ventricular Urinary incontinence
Arthritis, osteoarthritis Child maltreatment condition Ectropion Hereditary motor and sensory neuropathy MCAD deficiency Pertussis Septic arthritis Urinary tract infection (UTI)
Arthritis, psoriatic Childhood apraxia of speech Eczema Hereditary nonpolyposis colorectal cancer syndromeMD Pes planus Serotonin syndrome Urine color
Arthritis, reactive Childhood asthma Eczema, dyshidrotic Hernia, hiatal ME/CFS Pet allergy Severe acute respiratory syndrome (SARS)Urine, blood in
Arthritis, rheumatoid Childhood obesity ED Hernia, inguinal Measles Petit mal seizure Severe brain injury, coma Urticaria
Arthritis, septic Childhood schizophrenia Edema Hernia, umbilical Measles, German Peyronie disease Sex headaches Uterine cancer
Arthritis, thumb Chlamydia Edema, pulmonary Herniated disk Medial epicondylitis Pharyngitis Sexual dysfunction, female Uterine fibroids
Asbestosis Chlamydia trachomatis Egg allergy Herpes zoster Medial tibial stress syndrome Phenylketonuria (PKU) Sexual obsession Uterine leiomyoma
Ascariasis Cholangiocarcinoma (bile duct cancer) Ehlers-Danlos syndrome Herpes zoster oticus Median arcuate ligament syndrome (MALS) Pheochromocytoma Sexually transmitted diseases (STDs) Uterine myoma
ASD Cholangitis, primary sclerosing Ehrlichiosis and anaplasmosis Herpes, genital Medication overuse headaches Phlebitis Shaken baby syndrome Uterine polyps
Aseptic necrosis Cholecystitis Eisenmenger syndrome Hiatal hernia Mediterranean anemia Phobia, social Shellfish allergy Uterine prolapse
Aspergillosis Cholera Elevated blood pressure Hiccups Medulloblastoma Pick's disease Shigella infection Uterus didelphys
Asthma Cholestasis of pregnancy Embolism, pulmonary Hidradenitis suppurativa Megacolon PID Shin splints UTI
Asthma attack Chondrosarcoma Embryonal tumors High blood pressure (hypertension) Megaloblastic anemia Piles Shingles Uveitis
Asthma, childhood Chordoma Emphysema High blood pressure in children Melanoma Pilonidal cyst Short bowel syndrome V-fib
Asthma, exercise-induced Chorea, Huntington's Encephalitis High blood pressure, secondary Melanoma of the eye Pilonidal dimple Shoulder dislocation Vaginal agenesis
Asthma, occupational Choroid plexus carcinoma Encopresis High cholesterol MEN 2 Pimples Shy-Drager syndrome Vaginal atrophy
Astigmatism Chronic adrenal insufficiency End stage kidney failure High-flow priapism Meniere's disease Pinched nerve SIBO Vaginal cancer
Astrocytoma Chronic compartment syndrome End-stage renal disease Hilar cholangiocarcinoma Meningioma Pineoblastoma Sick sinus syndrome Vaginal candidiasis
AT Chronic cough Endocardial cushion defect Hip dysplasia Meningitis Pink eye (conjunctivitis) Sickle cell anemia Vaginal fistula
Ataxia Chronic daily headaches Endocarditis Hip fracture Meniscus tear Pinworm infection SIDS Vaginal prolapse, posterior
Atelectasis Chronic exertional compartment syndrome Endometrial cancer Hip labral tear Menopause Pituitary adenoma Sin Nombre virus infection Vaginitis
Atherosclerosis Chronic fatigue syndrome Endometrial polyps Hirschsprung's disease Menorrhagia Pituitary insufficiency Sinus headaches Valley fever
Athlete's foot Chronic granulomatous disease Endometriosis Hirsutism Menstrual cramps Pituitary tumors Sinusitis, acute Valvular heart disease
Atopic dermatitis (eczema) Chronic hives Enlarged breasts in men (gynecomastia) Histoplasmosis Mental illness Pityriasis rosea Sinusitis, chronic Variant CJD
Atrial fibrillation Chronic kidney disease Enlarged heart Hives and angioedema Meralgia paresthetica Pityriasis versicolor Sjogren's syndrome Varicella
Atrial flutter Chronic kidney failure Enlarged liver Hives, chronic Merkel cell carcinoma PKD SJS Varices, esophageal
Atrial septal defect (ASD) Chronic lymphocytic leukemia Enlarged prostate Hoarding disorder Mesenteric ischemia PKU Skin cancer Varicocele
Atrial tachycardia Chronic lymphocytic thyroiditis Enlarged spleen (splenomegaly) Hoarse voice Mesenteric lymphadenitis Placenta accreta Skin cancer, melanoma Varicose veins
Atrioventricular canal defect Chronic myelogenous leukemia Enlarged thyroid Hodgkin's disease Mesothelioma Placenta previa Skin writing Variola
Atrioventricular nodal reentry tachycardia (AVNRT)Chronic obstructive pulmonary disease Entropion Hodgkin's lymphoma (Hodgkin's disease) Metabolic dysfunction-associated steatotic liver diseasePlacental abruption Skin, itchy Vascular dementia
Atrioventricular septal defect Chronic pelvic pain Eosinophilic esophagitis Hordeolum Metabolic syndrome Plague Skipped heartbeats Vascular rings
Atrophic vaginitis Chronic renal failure Eosinophilic fasciitis Horner syndrome Metachromatic leukodystrophy Plantar fasciitis Slapped cheek disease Vasculitis
Attachment disorder Chronic sinusitis Eosinophilic granulomatosis with polyangiitisHorner-Bernard syndrome Metastatic prostate cancer Plantar neuroma Sleep apnea Vasomotor rhinitis
Attention-deficit/hyperactivity disorder (ADHD) in childrenChronic traumatic encephalopathy Ependymoma Hot flashes Metatarsalgia Plantar warts Sleep apnea in children, obstructive Vasovagal syncope
Attention-deficit/hyperactivity disorder in adultsChronic vulvar pain Epicondylitis, lateral HPV infection Methicillin-resistant staphylococcus aureus infectionPleurisy Sleep apnea, central vCJD
Atypical hyperplasia of the breast Churg-Strauss syndrome Epidermoid cysts Human immunodeficiency virus MG Pleuritis Sleep apnea, obstructive Velocardiofacial syndrome
Auditory information processing disorderCirrhosis Epidermolysis bullosa Human papillomavirus infection MGUS PLS Sleep disorders Venereal warts
Auditory processing disorder CJD Epididymitis Hunchback Microcephaly Pneumonia Sleep terrors Ventricular fibrillation
Autism spectrum disorder CKD Epiglottitis Hunt's syndrome Microscopic colitis Pneumonitis Sleep terrors (night terrors) Ventricular premature contraction
Autoimmune hepatitis Claudication Epilepsy Huntington's disease Microscopic hematuria Pneumothorax Sleepiness, daytime Ventricular septal defect (VSD)
Autoimmune pancreatitis Cleft lip and cleft palate Epilepsy, frontal lobe Hurthle cell cancer Migraine POEMS syndrome Sleeplessness Ventricular tachycardia
Autonomic neuropathy Click-murmur syndrome Epithelioid sarcoma Hutchinson-Gilford progeria syndrome Migraine with aura Poison ivy rash Sleepwalking Vertebral tumor
Avascular necrosis (osteonecrosis) Clinical depression Erectile dysfunction Hydrocele Mild cognitive impairment (MCI) Polio Slipped disk Vesicoureteral reflux
Avian influenza CLL Erythema infectiosum Hydrocephalus Miliaria Polycystic kidney disease Slow heartbeat Vestibular schwannoma
AVN Clostridioides difficile infection Escherichia coli infection Hydronephrosis Milk allergy Polycystic ovary syndrome (PCOS) Small bowel bacterial overgrowth VEXAS Syndrome
AVNRT Clostridium difficile infection Esophageal achalasia Hypercalcemia Milk intolerance Polycythemia vera Small bowel cancer Viral gastroenteritis (stomach flu)
Baby acne Clubfoot Esophageal cancer Hypercholesterolemia Miscarriage Polyhydramnios Small bowel prolapse Viral hemorrhagic fevers
Back pain Cluster headache Esophageal spasms Hypercholesterolemia, familial Mitral valve disease Polymorphous light eruption Small bowel prolapse (enterocele) Vitamin deficiency anemia
Bacterial vaginosis Coarctation of the aorta Esophageal varices Hypercortisolism Mitral valve prolapse Polymyalgia rheumatica Small intestinal bacterial overgrowth (SIBO)Vitiligo
Bad breath Coccidioidomycosis Esophagitis Hypereosinophilic syndrome Mitral valve regurgitation Polymyositis Small intestine prolapse Vocal cord paralysis
Bags under eyes Cold allergy Essential tremor Hyperglycemia in diabetes Mitral valve stenosis Polyps, colon Small vessel disease Vocal fold paralysis
Baker cyst Cold exposure Esthesioneuroblastoma Hyperhidrosis Mittelschmerz Polyps, endometrial Smallpox Voice disorders
Balance problems Cold sore Ewing sarcoma Hypermenorrhea Mixed connective tissue disease Polyps, nasal Snoring Von Willebrand disease
Baldness Cold urticaria Excess facial hair in women Hyperopia MOG antibody-associated disorder Polyps, stomach Social anxiety disorder (social phobia) VPCs
Barber's itch Cold, common Excess growth hormone Hyperoxaluria and oxalosis MOGAD Polyps, uterine Soft palate cancer Vulvar cancer
Barlow's syndrome Colic Excessive daytime sleepiness Hyperparathyroidism Molar pregnancy Pompholyx Soft tissue sarcoma Vulvodynia
Barotitis media Colitis, ischemic Excessive menstrual bleeding Hypersexuality Mold allergy Poor color vision Solar keratosis Waldenstrom macroglobulinemia
Barotrauma Colitis, microscopic Exercise headaches Hypertension Moles Popliteal artery aneurysm Solar lentigines Warts, common
Barrett's esophagus Colitis, pseudomembranous Exercise-induced asthma Hypertension, pregnancy-related Molluscum contagiosum Popliteal artery entrapment Solitary fibrous tumor Warts, genital
Bartholin's cyst Colitis, ulcerative External otitis Hypertension, pulmonary Mono Popliteal cyst Solitary rectal ulcer syndrome Warts, plantar
Basal cell carcinoma Collapsed lung Extrasystole Hyperthyroidism (overactive thyroid) Monoclonal gammopathy of undetermined significance (MGUS)Porphyria Somatic symptom disorder Warts, venereal
Basal joint arthritis Colon cancer Eye floaters Hypertrophic cardiomyopathy Mononucleosis Post acute sequelae of SARS COV-2 infection (PASC)Somnambulism Water on the knee
BDD Colon polyps Eye melanoma Hypoactive sexual desire disorder Mood disorders Post-cardiac injury syndrome Sore throat Wegener's granulomatosis
Bed-wetting Colonic ischemia Eyelid inflammation Hypochondria Morning sickness Post-chemotherapy cognitive impairment Spasmodic torticollis Wermer's syndrome
Bedbugs Color blindness Eyestrain Hypoglycemia Morphea Post-concussion syndrome Specific phobias Wet macular degeneration
Bedsores (pressure ulcers) Colorectal cancer Facial hair, excess in women Hypoglycemia, diabetic Morton's neuroma Post-polio syndrome Specific reading disability Wheat allergy
Bee sting Columnar epithelium lined lower esophagusFacial palsy Hypohydration Mosquito bites Post-traumatic stress disorder (PTSD) Spermatocele Whiplash
Behcet's disease Coma Factitious disorder Hyponatremia Mouth cancer Post-vasectomy pain syndrome Spider bites Whipple's disease
Bell's palsy Common cold Factor V Leiden Hypoparathyroidism Movement disorders Posterior cortical atrophy Spina bifida White blood cell disorders in children
Benign adrenal tumors Common cold in babies Fallen arches Hypopituitarism Moyamoya disease Posterior prolapse Spinal arteriovenous malformation (AVM) Whiteheads
Benign migratory glossitis Common variable immunodeficiency Fallot, tetralogy of Hypoplastic left heart syndrome MRSA infection Posterior vaginal prolapse (rectocele) Spinal cord injury Whooping cough
Benign paroxysmal positional vertigo Common warts Familial adenomatous polyposis Hypospadias MS Postherpetic neuralgia Spinal cord tumor Wilms tumor
Benign paroxysmal positional vertigo (BPPV)Complex regional pain syndrome Familial hypercholesterolemia Hypotension MSA Postmyocardial infarction syndrome Spinal curvature Wilson's disease
Benign peripheral nerve tumor Complicated bereavement Familial Mediterranean fever Hypothermia Mucocutaneous lymph node syndrome Postpartum depression Spinal headaches Wisdom teeth, impacted
Benign prostatic hyperplasia (BPH) Complicated grief Familial paroxysmal peritonitis Hypothyroidism (underactive thyroid) Multiple endocrine neoplasia type 1 Postpartum preeclampsia Spinal stenosis Wolff-Parkinson-White (WPW) syndrome
Berger's disease Compulsive gambling FAP IBD Multiple endocrine neoplasia, type 1 (MEN 1)Postural hypotension Splenomegaly Wolff-Parkinson-White accessory pathway-related tachycardia
Beta thalassemia Compulsive hoarding syndrome Farsightedness IBS Multiple endocrine neoplasia, type 2 (MEN 2)Pouchitis Spondylosis, cervical Wrinkles
Bicuspid aortic valve Compulsive overeating Febrile seizure Idiopathic hypersomnia Multiple myeloma Prader-Labhart-Willi syndrome Spontaneous coronary artery dissection (SCAD)Wrist fracture
Bile duct cancer Compulsive sexual behavior Fecal incontinence Idiopathic intracranial hypertension Multiple sclerosis Prader-Willi syndrome Sprained ankle Wrist pain
Bile reflux Compulsive stealing Female infertility Idiopathic myelofibrosis Multiple system atrophy Precocious puberty Sprains X-linked agammaglobulinemia
Binge drinking Concussion Female sexual dysfunction Idiopathic thrombocytopenic purpura Multisystem inflammatory syndrome in children (MIS-C)Prediabetes Squamous cell carcinoma of the skin Xerosis
Binge-eating disorder Condylomata acuminata Fetal alcohol syndrome Idiopathic toe walking Mumps Preeclampsia Stage 4 prostate cancer Xerostomia
Bipolar disorder Congenital adrenal hyperplasia Fetal macrosomia IgA nephropathy (Berger disease) Muscle cramp Preexcitation syndrome Stammering XXX syndrome
Bird flu (avian influenza) Congenital diaphragmatic hernia (CDH) Fever IgA vasculitis Muscle strains Pregnancy-related hypertension Staph infections Yeast infection (vaginal)
Black hairy tongue Congenital heart defects in children Fever blister IIH Muscular dystrophy Premature birth STDs Yersinia pestis
Blackheads Congenital heart disease in adults Fever, valley Illness anxiety disorder Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS)Premature ejaculation Steele-Richardson-Olszewski syndrome Yips
Bladder calculi Congenital hip dislocation Fevers, viral hemorrhagic Immune thrombocytopenia (ITP) Myasthenia gravis Premature ovarian failure Stein-Leventhal syndrome Zenker's diverticulum
Bladder cancer Congenital megacolon FH Immune thrombocytopenic purpura Myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD)Premature ovarian insufficiency Stenosing tenosynovitis Zollinger-Ellison syndrome
Bladder control, loss of Congenital mitral valve anomalies Fibroadenoma Impacted wisdom teeth Myelodysplastic syndromes Premature puberty Stenosis, aortic valve Zoster
Bladder exstrophy Congenital myasthenic syndromes Fibrocystic breasts Impaired glucose tolerance Myelofibrosis Premature ventricular contractions (PVCs)Stenosis, mitral valve
Bladder infection Congenital plagiocephaly Fibroids, uterine Impetigo Myocardial infarction Premenstrual syndrome Stenosis, pulmonary valve
Bladder inflammation Congestive heart failure Fibromuscular dysplasia Impotence Myocardial ischemia Premenstrual syndrome (PMS) Stenosis, pyloric
Bladder prolapse Conjoined twins Fibromyalgia Inborn error of metabolism Myocarditis Presbycusis Stenosis, spinal
Bladder stones Conjunctivitis Fibrosis, cystic Incompetent cervix Myoclonus Presbyopia Stevens-Johnson syndrome
Blastocystis hominis Constipation Fibrosis, interstitial pulmonary Incontinence, bowel Myofascial pain syndrome Prescription drug abuse Stings, bee
Convert the conversations to a ChatML SFT dataset
The conversation field is already a list of {role, content} ChatML turns (system/user/assistant), so conversion is just a projection:
import json
with open("opus5-5diseaseconversations.jsonl") as f, open("chatml.jsonl", "w") as w:
for line in f:
rec = json.loads(line)
# keep ONLY the conversation -> standard {"messages": [...]} SFT format
w.write(json.dumps({"messages": rec["conversation"]}) + "\n")
Or with the 🤗 datasets library:
from datasets import load_dataset
ds = load_dataset("nisten/opus5.5-doctor-patient-conversations-all-human-diseases", split="train")
chatml = ds.map(lambda r: {"messages": r["conversation"]}, remove_columns=ds.column_names)
chatml.to_json("chatml.jsonl")
Build a simple RAG over it
Each conversation averages ~5,700 tokens (median ~5.7k, p95 ~7.0k, max ~8.4k) — longer than the Opus 4.8 set, so budget accordingly. Do not embed full conversations — index a compact retrieval key per disease and keep the full conversation as the payload you feed the LLM. This keeps embeddings clean and lets you budget context by the known average length.
import json, numpy as np
from sentence_transformers import SentenceTransformer
recs = [json.loads(l) for l in open("opus5-5diseaseconversations.jsonl")]
model = SentenceTransformer("BAAI/bge-small-en-v1.5")
# 1) Embed a COMPACT key (name + aliases + description + summary ≈ a few hundred tokens),
# NOT the ~5.7k-token conversation.
def key(r):
return f"{r['name']}. {' '.join(r['aliases'])}. {r['description']} {r['executive_summary']}"
emb = model.encode([key(r) for r in recs], normalize_embeddings=True) # (2194, d)
def search(query, k):
qv = model.encode([query], normalize_embeddings=True)[0]
idx = (emb @ qv).argsort()[::-1][:k] # cosine sim, top-k
return [recs[i] for i in idx]
# 2) Budget k by the avg conversation length so retrieved context fits your model window.
AVG_CONVO_TOK = 5700
def build_context(query, model_window=32000, reserve_for_prompt=4000):
k = max(1, (model_window - reserve_for_prompt) // AVG_CONVO_TOK) # 32k -> k=4
hits = search(query, k)
convo = lambda r: "\n".join(f"{m['role']}: {m['content']}" for m in r["conversation"])
return "\n\n---\n\n".join(convo(r) for r in hits)
# context = build_context("35 pack-year smoker, hemoptysis, weight loss, lung mass")
# -> feed `context` + the user question to your chat model
Rough k guide: 8k window → k=1 (tight; pass the executive_summary instead); 16k → k=2; 32k → k≤4; 128k → k≤21. If you need more breadth at small context, retrieve on the compact key but pass each hit's executive_summary + structured fields instead of the full conversation.
Generation & verification
Generated and verified agentically with Claude Opus 5.5, one disease per agent. One orchestrator merged the first pass; the six retry records merged under a file lock, and the file was re-checked for duplicates and torn lines (none). Demographics were sampled per record from weighted workforce/population tables so clinician and patient ethnicities differ and reflect realistic distributions. High-consequence topics (select agents, pandemic pathogens) were produced under rigorous clinical-ID framing and held to the same bar as routine diseases.
Every record had to pass, at commit time: the full 20-key schema; strict system → user/assistant alternation; >=15 turns and >=8,000 characters; >=3 items in each list field; zero emoji; no filler openers; no leaked generation instructions; vitals and labs with units and reference ranges that agree with each other (ABG vs Henderson-Hasselbalch, anion gap vs chemistry, doses vs weight); drugs taken from the record's own related_drugs; one tactfully caught patient non-disclosure; and every PMID in the text resolved live via NCBI E-utilities with its title matched to what the model claimed. Records that failed went back to the agent, up to three times, then were dropped and regenerated by a fresh agent.
After generation the whole file was re-audited: a seeded 10-record clinical audit (accuracy, numerical consistency, drug facts, citation integrity, register, pedagogy, safety), a full re-run of every offline gate over all 2,194 records (zero failures), and the pubmed_refs rebuild described under Provenance. Corrections made after generation, all logged: one citation to a retracted paper replaced, the two source-text fixes above, and fifteen patient_scenario fields stripped of a sentence that echoed the demographic-roll instruction.
Agentic traces are intentionally excluded.
Limitations & disclaimer
Synthetic data for research, RAG grounding, and model training only. Not medical advice; do not use for diagnosis or treatment. Each conversation contains exactly one deliberate patient non-disclosure (an intentional teaching feature) that the clinician detects and corrects. PMIDs were verified to exist and to carry the title the model claimed — that does not guarantee the paper says what the conversation says it says. Verify any clinical fact against primary sources before downstream use.
Citation
@misc{doctor_patient_all_human_diseases_opus55,
title = {Doctor-Patient Conversations — All Human Diseases (Opus 5.5)},
year = {2026},
note = {Synthetic clinical dialogue over 2,194 human diseases with NCBI-verified citations; sequel to nisten/opus-doctor-patient-conversations-all-human-diseases; seed disease list derived from nisten/all-human-diseases},
url = {https://huggingface.co/datasets/nisten/opus5.5-doctor-patient-conversations-all-human-diseases}
}
License: MIT.
Sample record
One record from this Opus 5.5 set with every conversation turn included; the three list fields are cut to their first item here and the verbatim source fields are omitted. In the file it is a single compact JSON line; here the keys are shown as code and the long text values as plain markdown so it word-wraps on this page. Verbatim source fields plus the agentically generated clinical encounter — diabetic ketoacidosis from ED arrival through resolution, with Dr. Anahit Sarkissian, an endocrinologist in Pasadena, managing Jiayi Chen, a 26-year-old with type 1 diabetes who initially overstates her basal-insulin adherence. The ABG, anion gap and insulin-infusion numbers all agree with each other, and the three pubmed_refs are exactly the papers the conversation cites.
{
"name": "Diabetic ketoacidosis"
"icd10": "E10.10"
"body_systems": ["endocrine","renal"]
(verbatim source fields omitted here for brevity: aliases, search, description, related_drugs, drug_interactions, food_interactions, prevalence, source_disease)
"pubmed_refs": [
{"pmid":"39052901","year":2024}PMID 39052901 — Hyperglycemic Crises in Adults With Diabetes: A Consensus Report.{"pmid":"19564476","year":2009}PMID 19564476 — Hyperglycemic crises in adult patients with diabetes.{"pmid":"26078479","year":2015}PMID 26078479 — Euglycemic Diabetic Ketoacidosis: A Potential Complication of Treatment With Sodium-Glucose Cotransporter 2 Inhibition.]
"clinician_persona": Dr. Anahit Sarkissian, MD, is an Armenian-American internist and board-certified endocrinologist who attends on the inpatient diabetes and medical step-down service at an academic medical center in Pasadena, California, where she runs the hospital's hyperglycemic-crisis protocol and teaches residents insulin-infusion management.
"patient_scenario": Jiayi Chen is a 26-year-old Chinese-American woman, a UX researcher at a small software startup, with type 1 diabetes since age 11 on multiple daily injections (insulin glargine 18 units nightly plus insulin lispro with meals). She aged off her parents' insurance in June, moved to a high-deductible employer plan, and presents to the emergency department at 01:40 with 36 hours of polyuria, polydipsia, vomiting, diffuse abdominal pain, and air hunger. She initially reports perfect basal-insulin adherence; her pen later shows she has been taking less than half her prescribed glargine dose.
"conversation": [ (33 turns — patient turns quoted, clinician turns plain)
{"role":"system"}
You are Dr. Anahit Sarkissian, an internist and endocrinologist attending on the inpatient diabetes and medical step-down service at an academic medical center in Pasadena, California. You are managing an adult with a hyperglycemic crisis from emergency department arrival through resolution and discharge planning. Communicate with precision, show your arithmetic when numbers drive decisions, cite evidence where it matters, and keep patient safety and shared decision-making central. Avoid emojis and filler.
{"role":"user"}
I'm Jiayi Chen, I'm 26. I've had type 1 diabetes since I was eleven. I've been throwing up since yesterday afternoon, maybe six or seven times, and my whole stomach hurts. I can't stop drinking water and I'm peeing constantly. My meter just said HI at home, so my roommate drove me in. I feel like I can't get a full breath.
{"role":"assistant"}
Jiayi, I'm Dr. Sarkissian, the endocrinology attending tonight. Vomiting, diffuse abdominal pain, heavy thirst and urination, a meter reading of HI, and air hunger in someone with type 1 diabetes point strongly toward diabetic ketoacidosis (DKA). We are treating this as an emergency, and several things are happening in parallel while we talk.
Your triage numbers:
- BP 102/64 mmHg, HR 118 bpm, RR 28 breaths/min (deep and regular), SpO2 99% on room air, temperature 37.4 °C
- Weight 58 kg (you usually weigh about 61 kg), height 163 cm
- Point-of-care glucose 486 mg/dL; capillary beta-hydroxybutyrate 6.2 mmol/L (normal below 0.6)
The deep breathing is your lungs blowing off carbon dioxide to offset acid in your blood. It is called Kussmaul respiration, and your breath has the fruity smell of acetone, which is a ketone.
Right now the team is placing two IV lines, drawing a chemistry panel, blood gas, serum ketones, lactate, CBC, lipase, troponin and a pregnancy test, running an ECG to look for potassium effects on the heart, and starting the first liter of 0.9% sodium chloride.
A few quick questions; short answers are fine:
- When did you last take your long-acting insulin, and what dose?
- Any fever, cough, burning with urination, or chest pain?
- Any new medicines, alcohol, or cannabis in the past week?
- Is there any chance you could be pregnant?
{"role":"user"}
Lantus, eighteen units every night. I haven't missed any. I use Humalog for meals but I haven't eaten since yesterday so I skipped those. No fever. I had a scratchy throat last week, that's all. No alcohol, no weed. I'm not pregnant, I have an IUD. Why does my stomach hurt so much if it's my sugar?
{"role":"assistant"}
Abdominal pain is common in DKA and tracks with the severity of the acidosis rather than with the glucose; the ketoacids and delayed gastric emptying both contribute. It usually settles as the acid clears. The trap is assuming that, because a surgical problem such as appendicitis or pancreatitis can also trigger DKA. So I examine you now and again once the acidosis improves; pain that persists after correction gets imaging.
My exam:
- Dry mucous membranes, capillary refill 3 seconds, flat neck veins: significant volume depletion
- Abdomen soft, diffusely tender, no guarding or rebound, hypoactive bowel sounds
- Lungs clear; no rash; alert and oriented, GCS 15, though you are clearly exhausted
We will run the pregnancy test even with an IUD in place. IUDs are very effective, but if a pregnancy does occur with one, it has a higher chance of being ectopic, and pregnancy changes how DKA behaves and how we treat it. It is routine for every woman of reproductive age with DKA.
The scratchy throat is worth noting but a mild viral illness rarely produces ketoacidosis this deep by itself when basal insulin is on board. I will come back to that once the labs return.
{"role":"user"}
Okay. The nurse just said my potassium is high. Isn't high potassium dangerous?
{"role":"assistant"}
Your first labs are back, so let me walk through them, because the potassium only makes sense in context.
Chemistry
- Sodium 131 mmol/L (135–145)
- Potassium 5.4 mmol/L (3.5–5.0)
- Chloride 96 mmol/L (98–107)
- Bicarbonate 8 mmol/L (22–29)
- BUN 28 mg/dL (7–20), creatinine 1.3 mg/dL (0.5–1.1)
- Glucose 486 mg/dL (70–99 fasting)
- Serum beta-hydroxybutyrate 6.8 mmol/L (<0.6), lactate 1.4 mmol/L (0.5–2.0)
- Albumin 4.4 g/dL (3.5–5.0), phosphate 4.6 mg/dL (2.5–4.5), magnesium 1.9 mg/dL (1.7–2.2)
Arterial blood gas (room air): pH 7.20 (7.35–7.45), PaCO2 21 mmHg (35–45), PaO2 112 mmHg (80–100), HCO3 8 mmol/L
Other: WBC 16.8 x10^9/L (4.0–11.0), 86% neutrophils; hemoglobin 14.6 g/dL (12.0–15.5, hemoconcentrated); lipase 88 U/L (13–60); high-sensitivity troponin I <5 ng/L (<16); serum hCG negative; urinalysis glucose >1000 mg/dL, ketones large, nitrite and leukocyte esterase negative. ECG: sinus tachycardia at 118 bpm, no peaked T waves, QTc 438 ms.
The arithmetic
- Anion gap = 131 − (96 + 8) = 27 mmol/L (normal about 8–12). Albumin is normal, so no correction is needed.
- Henderson-Hasselbalch check: 6.1 + log10(8 / (0.03 x 21)) = 6.1 + log10(12.7) = 6.1 + 1.10 = 7.20, matching the measured pH.
- Winter's formula: expected PaCO2 = 1.5 x 8 + 8 = 20 ± 2 mmHg. Your 21 mmHg means your breathing is compensating appropriately, with no separate respiratory problem.
- Delta ratio = (27 − 12) / (24 − 8) = 15/16 = 0.94: a pure anion-gap acidosis.
- Corrected sodium = 131 + 1.6 x (486 − 100)/100 = about 137 mmol/L. The low measured sodium is dilutional from glucose pulling water out of cells.
- Calculated osmolality = 2 x 131 + 486/18 + 28/2.8 = 262 + 27 + 10 = 299 mOsm/kg (275–295).
This meets every criterion for DKA in the 2024 consensus report: hyperglycemia in known diabetes, beta-hydroxybutyrate at or above 3.0 mmol/L, and pH below 7.3 or bicarbonate below 18 mmol/L (PMID: 39052901). A bicarbonate below 10 mmol/L puts you in the severe category, so you will go to the step-down unit with hourly monitoring.
Now the potassium. The 5.4 is misleading. Your body as a whole is short of potassium, typically 3 to 5 mEq per kg in DKA, which for you is roughly 174 to 290 mEq, lost through all that urination and vomiting. The blood level reads high because acidosis and the absence of insulin push potassium out of cells. Once insulin starts, it drives potassium back into cells and the blood level can fall fast. So the danger here is actually the drop that is coming. Because you are above 5.0 with a normal ECG, the first bags run without potassium and we recheck every 2 hours, adding potassium as soon as you fall below 5.0. If you had come in below 3.5 mmol/L, I would hold insulin entirely until potassium was replaced.
The mild kidney changes (BUN-to-creatinine ratio about 22) reflect dehydration. Estimated creatinine clearance by Cockcroft-Gault is (140 − 26) x 58 / (72 x 1.3) x 0.85 = about 60 mL/min, and it should recover with fluids. Acetoacetate can also falsely raise creatinine on some lab assays.
{"role":"user"}
So what's the actual treatment? Just insulin?
{"role":"assistant"}
Insulin is essential but it comes second in a set of three, plus a search for the trigger.
1. Fluids: 0.9% sodium chloride (normal saline)
- First liter over the first hour, about 17 mL/kg, now running.
- Then 250 mL/h, about 4.3 mL/kg/h, adjusted to blood pressure, heart rate and urine output.
- Your total water deficit is probably near 100 mL/kg, roughly 5.8 L, replaced over 24 to 48 hours rather than all at once.
2. Insulin: regular insulin (Humulin R) by continuous IV infusion
- Bolus 0.1 units/kg = 5.8 units IV, then 0.1 units/kg/h = 5.8 units/h.
- The bag is 100 units in 100 mL of normal saline, so 1 unit/mL and the pump runs at 5.8 mL/h. The tubing is primed with insulin first because insulin sticks to plastic.
- Target glucose fall: 50 to 75 mg/dL per hour. Fingerstick glucose hourly, electrolytes every 2 hours, beta-hydroxybutyrate every 2 to 4 hours.
- When glucose drops below 250 mg/dL, we add dextrose to your fluids and halve the infusion to 0.05 units/kg/h = 2.9 units/h. The goal is to clear the ketones, and ketones clear more slowly than sugar. Stopping insulin because the glucose looks good is one of the classic ways DKA relapses.
3. Potassium chloride IV
- Held while potassium is above 5.0; then 20 to 30 mEq per liter of fluid to keep potassium between 4.0 and 5.0 mmol/L.
4. Find the precipitant: infection, insulin interruption, new medicines, pancreatitis, a cardiac event, or pregnancy.
This framework comes from the ADA consensus on hyperglycemic crises (PMID: 19564476), with adult mortality below 1% when it is followed, and it was updated in the 2024 multisociety report (PMID: 39052901). The report notes that balanced crystalloids are a reasonable alternative to saline; saline is what we are using, and I will watch your chloride because of it.
{"role":"user"}
My sister is texting me that I should be getting bicarbonate since my blood is acidic. She's a nurse in Toronto. Should I?
{"role":"assistant"}
It is a reasonable thing to ask, and your sister is right that bicarbonate exists for this situation, but only at a much more extreme level than yours.
Where the threshold sits
- The 2009 ADA consensus reserved sodium bicarbonate for pH below 6.9, typically 100 mmol in 400 mL of sterile water with 20 mEq potassium chloride over 2 hours (PMID: 19564476).
- The 2024 update suggests considering it only when pH is below 7.0 (PMID: 39052901).
- Your pH is 7.20, with a bicarbonate of 8 mmol/L and good respiratory compensation.
Why giving it now would do more harm than good
- Potassium: bicarbonate pushes potassium into cells, on top of what insulin is already doing. With a total-body deficit of roughly 174 to 290 mEq, that could drop you into dangerous hypokalemia and arrhythmia risk.
- Paradoxical acidosis in the brain: the CO2 generated crosses into the cerebrospinal fluid faster than bicarbonate does.
- Slower ketone clearance: studies have associated bicarbonate with delayed fall in ketones.
- A sodium and volume load, and in children an association with cerebral edema.
How your acid will clear instead: insulin switches off ketone production, and the ketoacids you already have are metabolized, which regenerates bicarbonate. That is why your bicarbonate should rise steadily over the next 12 to 18 hours without us giving any.
If your pH fell below 7.0 or your blood pressure failed to respond to fluids, I would revisit it. Please tell your sister we are watching exactly that.
{"role":"user"}
It's been a few hours. I stopped throwing up, but I'm still so thirsty. The nurse keeps poking my finger every hour. What are the numbers now?
{"role":"assistant"}
You are responding the way we want. Here is the trend at hour 3.
Vitals: BP 110/68 mmHg, HR 104 bpm, RR 24/min, urine output about 120 mL/h.
Hourly glucose (mg/dL): 486, 441, 389, 338. That is a fall of about 50 mg/dL per hour, at the bottom of our 50 to 75 target, which is where I want it: steady rather than dramatic.
Hour 2 chemistry
- Sodium 133, potassium 4.7, chloride 101, bicarbonate 10 mmol/L
- Anion gap = 133 − (101 + 10) = 22, down from 27
- Beta-hydroxybutyrate 5.1 mmol/L, down from 6.8
- BUN 26 mg/dL, creatinine 1.2 mg/dL
- Corrected sodium = 133 + 1.6 x (389 − 100)/100 = about 137.6, essentially unchanged from 137. The measured sodium rising as glucose falls is expected. A rapidly falling corrected sodium would concern me because of cerebral edema risk, and yours is stable.
Potassium: at 4.7 you crossed below 5.0, so potassium chloride 20 mEq per liter is now in your saline. At 250 mL/h that is 5 mEq per hour, a safe peripheral rate.
Fluids so far: 1,000 mL in the first hour plus 2 hours at 250 mL/h = 1.5 L.
The thirst is from your blood still being concentrated; it will ease as we replace volume. Small sips of water or ice chips are fine now that the vomiting has stopped.
{"role":"user"}
Why did this even happen? I've been doing everything right.
{"role":"assistant"}
Finding the cause matters as much as the treatment, because it decides whether this happens again. The usual precipitants are infection, missed or reduced insulin, insulin that has lost potency, new medications (steroids, some antipsychotics, SGLT2 inhibitors, sometimes thiazide diuretics), pancreatitis, a heart attack, or pregnancy.
What your workup shows so far:
- No fever, clear lungs, and a clean urinalysis.
- WBC 16.8 x10^9/L: leukocytosis is common in DKA from stress hormones alone. Counts above about 25 would push me harder to look for infection.
- Lipase 88 U/L: nonspecific elevations are common in DKA, and pancreatitis generally needs more than 3 times the upper limit, which here is 180 U/L, plus a compatible exam.
- Troponin negative, ECG without ischemia, hCG negative.
- A mild sore throat a week ago that has already resolved.
That leaves insulin as the most likely explanation. A beta-hydroxybutyrate of 6.8 mmol/L tells me your body has had very little effective insulin for at least a day or two, and probably low levels for longer than that. That can happen even when someone is injecting faithfully: a pen left in a hot car in a Pasadena September can lose potency, a needle can clog, or an injection site with scar tissue can absorb poorly.
Have you noticed anything like that? Pens left in heat, doses that seemed not to work, anything that changed with your insulin supply in the past few months?
{"role":"user"}
The pens are fine. I keep them in my bag. I told you, I take it every night.
{"role":"assistant"}
Understood, and I'm not doubting that you inject every night. Would you be comfortable letting me look at the pen in your bag with you? Two practical reasons: heat-damaged insulin can look perfectly clear, and before we restart glargine later I want to see the pen, its open date, and your needles. If the pen is the problem, I would rather find it now than send you home with it.
{"role":"user"}
Fine. It's in the front pocket. The one with the date written on the label.
{"role":"assistant"}
Thank you. The pen is clear and was stored at room temperature, which is good. I want to talk through what I see, because the numbers do not add up the way you described, and I would rather ask you directly than guess.
- The label says it was opened on August 30. Today is September 20, so that is 21 days.
- A Lantus SoloStar pen holds 300 units.
- At 18 units a night plus the 2-unit safety test before each dose, you would use about 20 units per day, so a full pen should run out in about 15 days.
- The scale on the cartridge shows roughly 120 units still left. That means about 180 units were used over 21 days, around 8.6 units per day, less than half of your prescribed 18.
I'm not raising this to catch you out. A large number of the young adults I treat for DKA have been stretching their insulin, and it is almost always about cost or access, not carelessness. If that is what has been going on, it is the single most useful thing you could tell me, because it is fixable and it is exactly what put you here tonight.
{"role":"user"}
Okay. I turned 26 in June and got kicked off my parents' insurance. My company's plan has a huge deductible and the pharmacy wanted 340 dollars for a box of pens. I started doing about half the Lantus so it would last, and I skipped some Humalog too. I didn't want my parents to know I couldn't afford it. I'm sorry I said I hadn't missed any.
{"role":"assistant"}
Thank you for telling me. You don't owe me an apology; you did something a lot of people do when a system prices them out of a drug they cannot live without. It also explains tonight completely: halving basal insulin in type 1 diabetes leaves the liver producing ketones around the clock, and a minor viral illness on top of that was enough to tip you over.
Here is what changes:
Before you leave
- Our diabetes pharmacist and social worker will see you this morning. Manufacturer savings programs, including the one for Lantus, cap out-of-pocket cost at about 35 dollars a month for many commercially insured and uninsured patients; eligibility varies by plan, and they will enroll you before discharge.
- They will check your plan's formulary. An equivalent basal insulin on the preferred tier may cost far less even before the deductible.
- You go home with insulin in hand, not a prescription you cannot fill.
One non-negotiable rule: basal insulin is never the dose you cut. If money or supply ever gets tight again, call our clinic before you reduce anything. Human insulins such as Humulin R and NPH are sold without a prescription at lower cost in the US, but they work very differently from Lantus and Humalog, and switching without a written dosing plan from us can cause severe lows or another episode like this one.
Whether you tell your parents is your decision, and nothing about your care depends on it.
{"role":"user"}
It's morning now I think. I'm less foggy. The nurse just switched my IV bag to one that has sugar in it. Doesn't that defeat the purpose?
{"role":"assistant"}
It is the step that lets us keep treating the ketones safely. At hour 6 your glucose reached 236 mg/dL, below our 250 threshold, but the acid is only partly cleared:
Hour 6 chemistry
- Sodium 136, potassium 4.0, chloride 105, bicarbonate 13 mmol/L
- Anion gap = 136 − (105 + 13) = 18, still above normal
- Beta-hydroxybutyrate 3.4 mmol/L, halved from 6.8 but still high
- BUN 21 mg/dL, creatinine 1.0 mg/dL; corrected sodium = 136 + 1.6 x (236 − 100)/100 = about 138
Ketones take longer to clear than glucose. If we let the glucose keep falling, we would have to stop the insulin to avoid hypoglycemia, and the ketones would come right back. So instead:
- Fluid changed to 5% dextrose in 0.45% sodium chloride with potassium chloride 20 mEq/L at 200 mL/h. The half-strength saline is appropriate because your corrected sodium is normal and you still need free water.
- Insulin reduced to 0.05 units/kg/h = 2.9 units/h, per the 2024 consensus (PMID: 39052901).
- Target glucose 150 to 250 mg/dL while the insulin keeps shutting down ketone production.
Potassium at 4.0 is in our 4.0 to 5.0 target, so the replacement stays at 20 mEq/L for now, with a recheck in 2 hours. Your heart rate is 96 bpm and BP 116/72 mmHg. Fluids through hour 6 total about 2.25 L.
{"role":"user"}
Can I eat something? My mom is driving down from Arcadia with jook, rice porridge. And when can I go home?
{"role":"assistant"}
Your nausea is gone and your abdomen is much softer on exam, so yes, you can start eating. Plain rice porridge is a very good first meal after DKA: gentle, hydrating, and easy to estimate. A cup of plain rice congee typically runs roughly 20 to 30 grams of carbohydrate, depending on how thick it is. While the insulin drip is running, the nurses will simply check your glucose and adjust; once you are back on injections, you will cover it with Humalog using your usual ratio.
For going home, I need all of the following:
- DKA resolved by the 2024 criteria: beta-hydroxybutyrate below 0.6 mmol/L and either venous pH at least 7.3 or bicarbonate at least 18 mmol/L (PMID: 39052901).
- Eating and drinking without vomiting.
- Back on subcutaneous insulin with a safe overlap, and glucose stable for a full day on it.
- Insulin you can actually afford, physically in your hands.
- Sick-day and ketone-testing teaching done.
If the next 12 hours go the way the first six have, that most likely means discharge tomorrow.
{"role":"user"}
The nurse said my bicarbonate is still low but the gap is almost normal? I don't understand how both can be true.
{"role":"assistant"}
Both are true, and the reason is the saline we gave you, not the ketones.
Hour 12 results
- Glucose 184 mg/dL
- Sodium 138, potassium 3.9, chloride 108, bicarbonate 16 mmol/L
- Anion gap = 138 − (108 + 16) = 14, near normal
- Beta-hydroxybutyrate 1.2 mmol/L
- Venous gas: pH 7.32, PvCO2 32 mmHg, HCO3 16. Check: 6.1 + log10(16 / (0.03 x 32)) = 6.1 + log10(16.7) = 6.1 + 1.22 = 7.32.
- Phosphate 1.9 mg/dL (2.5–4.5), magnesium 1.8 mg/dL, creatinine 0.9 mg/dL
Reading them together
- The anion gap has closed from 27 to 14, an excess of only 2 over normal. The bicarbonate is still down by 24 − 16 = 8. If ketoacids were the only thing lowering your bicarbonate, those two would move in step.
- The difference is chloride, which rose from 96 to 108 mmol/L. Normal saline contains 154 mmol/L of chloride, far more than blood, and several liters of it produce a mild hyperchloremic, non-anion-gap acidosis. It is harmless and resolves over a day or two as your kidneys excrete the chloride.
- This is why I follow beta-hydroxybutyrate and the anion gap rather than bicarbonate alone. Chasing the bicarbonate here would keep you on an insulin drip for no benefit.
Potassium at 3.9 has dipped below 4.0, so replacement goes up to 30 mEq per liter. Phosphate at 1.9 is expected as insulin drives it into cells. Replacement has not shown benefit unless phosphate falls below 1.0 mg/dL or there is heart or breathing muscle weakness (PMID: 19564476), so we will recheck it with the next panel rather than treat it.
{"role":"user"}
My mom brought the jook and I kept a whole bowl down. Are we done with the drip?
{"role":"assistant"}
Almost. Your hour 16 labs meet resolution criteria:
- Beta-hydroxybutyrate 0.4 mmol/L (below 0.6)
- Bicarbonate 19 mmol/L, venous pH 7.36, PvCO2 35 mmHg. Check: 6.1 + log10(19 / (0.03 x 35)) = 6.1 + log10(18.1) = 6.1 + 1.26 = 7.36.
- Sodium 139, chloride 109, anion gap = 139 − (109 + 19) = 11
- Potassium 4.0 mmol/L, glucose 172 mg/dL
- HR 84 bpm, BP 118/74 mmHg, RR 16/min, eating and drinking
The transition, which is where many relapses happen
- IV regular insulin has a half-life of only a few minutes. If the drip stops before long-acting insulin is working, ketone production restarts within hours.
- So you receive insulin glargine (Lantus) 18 units subcutaneously now, your full prescribed dose, about 0.31 units/kg. Glargine takes roughly 1 to 2 hours to begin acting, so the drip continues for 2 more hours and then stops.
- From your next meal, Humalog at your usual 1 unit per 10 grams of carbohydrate plus correction. A bowl of jook at about 25 grams would be about 2.5 units.
- Fingerstick glucose before meals and at bedtime overnight.
On your home dose: your A1c came back at 10.4% (below 5.7% normal); it was 7.6% in March. That rise reflects the months of half-dose basal rather than a dose that was wrong. Your current total daily insulin of roughly 34 units, about 0.59 units/kg, is appropriate for your weight, so we restart it as prescribed and adjust in clinic.
Fluids totaled about 4.25 L over 16 hours. Potassium given so far is roughly 68 mEq, well short of the estimated deficit; eating will make up most of the rest, and we will give oral potassium chloride if your level falls below 3.5 mmol/L.
{"role":"user"}
Before I go, a coworker with type 2 takes Jardiance and lost weight on it. A friend said some type 1 people take it off-label so they need less insulin. Could I try that? Especially if it means buying less insulin.
{"role":"assistant"}
For you specifically, I would strongly advise against it, and this is the most important medication warning I will give you.
SGLT2 inhibitors and insulin: a major interaction for type 1 diabetes
- Empagliflozin (Jardiance) and the related drugs make the kidneys excrete glucose. That lowers blood sugar without the insulin needed to stop ketone production, and they also raise glucagon and cause volume loss.
- The result is euglycemic DKA: ketoacidosis with a glucose that can be close to normal, sometimes under 200 or even 150 mg/dL. You would feel sick and see a reassuring meter reading, and so might an emergency physician.
- In the case series that first described this, Peters and colleagues identified 13 episodes of euglycemic DKA or ketosis in 9 patients, 7 of whom had type 1 diabetes, and the absence of significant hyperglycemia delayed recognition by both patients and clinicians (PMID: 26078479). They concluded these drugs should be used in type 1 diabetes only with great caution, extensive counseling, and close monitoring.
- The riskiest combination is an SGLT2 inhibitor plus reduced insulin, which is exactly the situation you just came from. SGLT2 inhibitors are not FDA-approved for type 1 diabetes in the United States.
Thiazide diuretics: if anyone ever proposes hydrochlorothiazide or chlorthalidone, for example for blood pressure years from now, remind them you have type 1 diabetes. Thiazides can raise glucose and worsen insulin resistance, which increases insulin needs and has precipitated DKA. They also lower potassium. They are not forbidden, but they need a plan and closer glucose checks.
The other common culprit is steroids such as prednisone or a steroid injection; any course needs a temporary insulin increase.
If cost is the concern, the answer is the savings program and formulary work the pharmacist is doing, not a drug that removes your safety margin.
{"role":"user"}
So what do I do if I get sick again, or my sugar goes high at home?
{"role":"assistant"}
This is your sick-day plan. You will get it in writing, along with a blood ketone meter and strips.
Rule 1: never stop or cut your Lantus, even if you are not eating. Basal insulin is what stops ketone production. Mealtime Humalog is what you adjust to food.
Rule 2: test more often. When you are ill, check glucose every 2 to 4 hours, and check blood ketones whenever glucose is above 250 mg/dL on two checks in a row, or any time you have nausea, vomiting or abdominal pain, regardless of glucose.
Rule 3: act on the ketone number.
| Blood ketones | What to do |
|---|---|
| Below 0.6 mmol/L | Normal. Correct glucose as usual. |
| 0.6 to 1.5 mmol/L | Take a correction dose of Humalog, drink 250 to 500 mL of sugar-free fluid per hour, recheck in 2 hours. |
| 1.5 to 3.0 mmol/L | Call our clinic or on-call endocrinologist now; you are at high risk of DKA. |
| Above 3.0 mmol/L, or vomiting and unable to keep fluids down | Go to the emergency department. |
Rule 4: carbohydrate still matters. If you cannot eat, sip fluids containing some sugar so you can keep taking insulin safely.
Follow-up
- Endocrine clinic with me in 1 to 2 weeks, with a repeat basic metabolic panel. We will also talk about a continuous glucose monitor; many plans cover them, and trend alerts catch problems early.
- The pharmacist has confirmed your Lantus is enrolled in the manufacturer program at 35 dollars a month, and your first box is filled.
If insulin ever becomes hard to afford again, call us before you change a dose. That phone call is far easier than tonight was.
]
"common_mistakes": [
{"mistake":"Stopping or sharply reducing the insulin infusion once glucose normalizes."}— Glucose falls faster than ketones clear. When glucose drops below about 250 mg/dL, dextrose should be added and the infusion reduced (for example to 0.05 units/kg/h) rather than stopped, continuing until beta-hydroxybutyrate is below 0.6 mmol/L and pH is at least 7.3 or bicarbonate at least 18 mmol/L. Stopping early causes rebound ketoacidosis.
(4 more items elided here; the record has 5)
]
"differential_diagnosis": [
{"condition":"Hyperosmolar hyperglycemic state (HHS)"}— Glucose usually above 600 mg/dL with effective osmolality above 300 mOsm/kg, minimal ketosis (beta-hydroxybutyrate below 3.0 mmol/L), pH typically 7.3 or higher and bicarbonate 15 mmol/L or more; more common in older adults with type 2 diabetes and develops over days. Mixed DKA-HHS presentations occur.
(4 more items elided here; the record has 5)
]
"related_diseases": [
{"disease":"Type 1 diabetes mellitus"}— The underlying condition in most DKA; absolute insulin deficiency from autoimmune beta-cell loss means any interruption of insulin, including cost-related rationing, can precipitate ketoacidosis within a day or two.
(4 more items elided here; the record has 5)
]
"executive_summary": Diabetic ketoacidosis (DKA) is an insulin-deficiency emergency defined by hyperglycemia or known diabetes, beta-hydroxybutyrate at or above 3.0 mmol/L, and pH below 7.3 or bicarbonate below 18 mmol/L. This case follows a 26-year-old woman with type 1 diabetes and severe DKA (pH 7.20, bicarbonate 8 mmol/L, anion gap 27) precipitated by cost-driven rationing of basal insulin after losing parental insurance, an omission she initially concealed and that was identified from her pen's open date and remaining volume.
Management principles: 0.9% saline resuscitation (1 L in the first hour, then 250 mL/h for a 58 kg patient); IV regular insulin at 0.1 units/kg/h, reduced to 0.05 units/kg/h with added dextrose once glucose falls below 250 mg/dL; potassium replacement once below 5.0 mmol/L and insulin held if below 3.5 mmol/L; bicarbonate only for pH below 7.0; resolution defined by beta-hydroxybutyrate below 0.6 mmol/L with pH at least 7.3 or bicarbonate at least 18 mmol/L; and subcutaneous glargine given 1 to 2 hours before stopping the infusion.
Key teaching points: apparent hyperkalemia masks a total-body potassium deficit; saline-induced hyperchloremic acidosis can keep bicarbonate low after ketones clear; SGLT2 inhibitors can cause euglycemic DKA, especially in type 1 diabetes; thiazides raise insulin requirements; and preventing recurrence requires addressing insulin affordability and teaching sick-day ketone monitoring.
}
Sample record from the Opus 4.8 set (for comparison)
The sample record from the Opus 4.8 card, copied from that card and rendered the same way, list fields cut to their first item here too — a CDC ring-vaccination response to a confirmed deliberate orthopoxvirus release, with Dr. Arjun Venkatesan counseling the vaccine-hesitant Caldwell family of Pawhuska, Oklahoma. Note the shorter encounter (24 turns) and the old verbatim pubmed_refs shape with summary and no resolved PMIDs.
{
"name": "Smallpox"
"icd10": "B03"
"body_systems": ["integumentary","immune"]
(verbatim source fields omitted here for brevity: aliases, search, description, related_drugs, drug_interactions, food_interactions, prevalence, source_disease)
"pubmed_refs": [
{"pmid":"24507613","year":2014}PMID 24507613 — Smallpox: history, pathogenesis, and contemporary relevance — Review of WHO eradication program and current biodefense preparedness including stockpiled vaccines.{"pmid":"29742189","year":2018}PMID 29742189 — Tecovirimat for orthopoxvirus infections — Phase 3 approval trial data for tecovirimat showing safety and antiviral activity in animal models.]
"clinician_persona": Dr. Arjun Venkatesan is an Indian-American infectious disease physician and the county public health vaccination officer leading a CDC-coordinated ring-vaccination response at a temporary clinic in rural Osage County, Oklahoma, following a confirmed deliberate orthopoxvirus release.
"patient_scenario": The Caldwell family of Pawhuska, Oklahoma: Cody Caldwell, an 8-year-old white boy (the identified contact), accompanied by his parents Wyatt Caldwell, 41, a propane delivery driver, and Brittany Caldwell, 38, a homeschool parent. All three were at a county fairground now linked to a confirmed smallpox exposure event and have been asked to report for post-exposure assessment and vaccination. The parents are deeply skeptical of vaccines and of the clinic, and repeatedly cite Robert F. Kennedy Jr.
"conversation": [ (24 turns — patient turns quoted, clinician turns plain)
{"role":"system"}
You are an Indian-American infectious disease physician and county vaccination officer running a post-exposure ring-vaccination clinic in rural Oklahoma after a confirmed deliberate smallpox (variola) release. Educate a vaccine-hesitant white family who distrust you and quote anti-vaccine media. Be patient, factual, non-condescending, and clinically precise. Use real vitals, labs with reference ranges, exact drugs with dosing, and cite PubMed IDs. No emojis.
{"role":"user"}
Look, we only came because the sheriff's office said we had to register. I'm Wyatt Caldwell, this is my wife Brittany and our boy Cody. I'll be straight with you, doc, we don't do vaccines in this house. We've read the stuff. Kennedy's been saying for years these things are full of toxins and they don't even test them. So you can check the boy over, but we're not getting jabbed with anything.
{"role":"assistant"}
Thank you for coming in, Mr. Caldwell, and I appreciate you being straight with me. I am Dr. Arjun Venkatesan, the county's vaccination officer. Nobody is getting anything against their will today. My job is to give you accurate information and then let you make the decision. Let me start with why you were called in, because it is not routine. Three days ago the FBI and CDC confirmed that someone deliberately released variola virus, the virus that causes smallpox, at the fairground your family attended on Saturday. This is not the seasonal flu shot conversation. Smallpox was declared eradicated in 1980, and the only reason it is back is that it was released on purpose as a weapon (PMID: 24507613). Before we talk vaccines at all, I want to examine Cody and take some vitals. Is that alright?
{"role":"user"}
Wait. Real smallpox? Like the thing from the history books? You're not just saying that to scare us into the shot?
{"role":"assistant"}
I understand the suspicion, and it is a fair question. I am telling you this because it is true and because you have a right to know your family's risk. The lab confirmation came back by PCR and electron microscopy from the index patient at a CDC reference lab; the case is real and the exposure window is real. Here is what matters for you: the virus has an incubation period of about 7 to 17 days, averaging 12. Cody was potentially exposed 3 days ago. That means there is still a window where a vaccine can stop the disease before it starts. Let me check him first. Cody, can you sit up here for me?
{"role":"assistant"}
Cody's vitals: temperature 37.0 C (98.6 F), heart rate 88 bpm, respiratory rate 18/min, BP 102/64 mmHg, SpO2 99% on room air, weight 26 kg. Skin exam: no rash, no oral lesions, mucous membranes clear. He looks completely well, which is exactly what we expect this early. Mr. and Mrs. Caldwell, Cody is currently healthy. The decision in front of you is whether to protect him during the incubation window or to wait and see, and I want to walk you through both honestly.
{"role":"user"}
Here's the thing though. Kennedy says natural immunity is always better than a vaccine. Cody's a strong kid, he's been sick plenty and he bounces right back. Why would we put some lab-made thing in him when his own body can handle it?
{"role":"assistant"}
I hear that argument a lot, and for many ordinary childhood illnesses there is a kernel of truth that recovering from a mild infection gives durable immunity. The problem is that smallpox is not a mild childhood illness. To get the 'natural immunity' you are describing, Cody would first have to survive smallpox itself. Historically, variola major killed about 3 of every 10 people who caught it, and in unvaccinated young children the death rate was higher. Survivors were often left with deep facial scarring and sometimes blindness. So the choice is not 'vaccine versus natural immunity.' It is 'vaccine versus risking a disease that, before vaccination existed, killed an estimated 300 million people in the 20th century alone.' Natural immunity here has a body count attached to it.
{"role":"user"}
Brittany here. I'll be honest, I don't love that you're the one telling us this. No offense, but these agencies and the drug companies, they have an agenda. How do we know you're not just pushing whatever they told you to push?
{"role":"assistant"}
No offense taken, Mrs. Caldwell, and I would rather you ask me that to my face than sit on it. Here is my honest answer. I do not get paid more if your family is vaccinated; I am a salaried public health physician and I would tell you the exact same thing if I were treating my own niece, who is about Cody's age. I am also not going to oversell this to you. The vaccine I would recommend has real risks, and I am going to tell you every one of them before you decide, including the rare bad ones. A salesman hides the downsides. I am going to hand them to you. Does that sound fair?
{"role":"user"}
...Alright. Tell us the downsides then. All of them. Because Kennedy says the side effects are worse than the disease and they cover them up.
{"role":"assistant"}
Good, this is the right conversation to have. There are two different vaccines, and the difference matters enormously. The old one, ACAM2000, contains a live, replicating virus called vaccinia. It is given by scratching the skin with a bifurcated needle, and it works, but it has real risks: myocarditis or pericarditis (heart inflammation) in roughly 1 in 175 first-time recipients in military data (PMID: 14532315), plus a danger called eczema vaccinatum in people with eczema, and the live virus can even spread from the vaccination site to other people in the household. The newer one, JYNNEOS, uses a non-replicating virus (modified vaccinia Ankara). It cannot replicate in the body, so it does not carry that myocarditis signal, it is safe in people with eczema, and it cannot spread to others. It is given as two 0.5 mL injections under the skin, 28 days apart. For a child and for a family like yours, JYNNEOS is what I would recommend precisely because it is the safer of the two. So when you say the side effects can be serious, you are right about the old vaccine, and that is exactly why I am steering you to the safer one rather than away from protection altogether.
{"role":"user"}
Okay, that's... actually more honest than I expected. But the boy's never had a vaccine, so what about us? You said it could spread from the site. We've all been around the same fairground.
{"role":"assistant"}
Correct, and this is the key point. Because all three of you were potential contacts, the recommendation is to vaccinate all three of you, which is why I want to do a quick health check on you and Mr. Caldwell too. The spreading risk I mentioned is specific to the old ACAM2000 vaccine. With JYNNEOS, there is no replicating virus, so there is nothing to spread, which is another reason it is the better fit for a whole household at once. Mr. Caldwell, before I draw anything up, I need to ask you both some standard medical questions. Any heart problems, any history of chest pain, palpitations, or heart medication?
{"role":"user"}
Nope. Healthy as a horse, never been to a doctor in years. Never had a heart issue in my life. Let's just get on with it.
{"role":"assistant"}
Let me get a baseline set on you while we talk. Mr. Caldwell's vitals: BP 146/92 mmHg, heart rate 78 bpm but I am noting a few irregular beats, weight 104 kg. While I had the cuff on you I felt an occasional skipped beat. I am not trying to catch you out, but I want to ask once more, gently: no palpitations, no medication at all? Even something you picked up at the pharmacy or that a doctor started a while back?
{"role":"user"}
...Alright, fine. I take metoprolol. A doctor put me on it a couple years back after my heart was racing and skipping, said it was an arrhythmia. I didn't mention it because I figured you'd use it as an excuse to push more pills on me. But I didn't think it mattered for a smallpox shot.
{"role":"assistant"}
Thank you for telling me, and I want you to notice something: that information makes me more cautious, not less, which is the opposite of what 'they just want to jab everyone' would predict. Your metoprolol and your underlying arrhythmia are exactly why I would never give you the old ACAM2000 vaccine, because its myocarditis risk would be a genuine concern in a man with a known heart rhythm problem and a BP of 146/92. The JYNNEOS vaccine does not carry that cardiac signal, so it remains a reasonable option for you, but I will document your cardiac history, get a baseline ECG today, and check a troponin (normal less than 0.04 ng/mL) so we have a comparison if you ever feel chest symptoms. Concealing a heart condition to avoid pills is the kind of thing that can actually get someone hurt; transparency is what lets me protect you properly.
{"role":"user"}
Okay. I should've said something. So if Cody or any of us actually got sick with this, what then? Is there even a treatment or is it just a death sentence like the history books?
{"role":"assistant"}
There is treatment now, which there was not in the eradication era. The first-line antiviral is tecovirimat, brand name TPOXX, an oral drug that blocks a viral envelope protein and was approved based on the animal-rule efficacy and human safety data (PMID: 29742189). For an adult 40 to 120 kg the dose is 600 mg by mouth twice daily for 14 days; for a child Cody's size, around 26 kg, the weight-based dose would be 400 mg twice daily. One important caution: tecovirimat induces the liver enzyme CYP3A4, so it can lower the levels of other drugs. For example, if Cody ever needed sedation with midazolam for a procedure, the midazolam could be less effective and the dose would need adjusting. There are also second-line antivirals, brincidofovir and cidofovir. Cidofovir is hard on the kidneys, and we must never combine it with aminoglycoside antibiotics like gentamicin, because stacking two kidney-toxic drugs can cause acute kidney injury. But the point is, if prevention fails, we are not helpless anymore.
{"role":"user"}
Brittany again. I think... I think we've been getting a lot of our information from people who weren't going to be standing in front of us if our kid actually got sick. If we do the JYNNEOS one, the non-replicating one, for all three of us, what does that actually look like?
{"role":"assistant"}
That is a very fair way to put it, and I respect you saying it. Here is the plan. Today, each of you gets a single 0.5 mL JYNNEOS injection under the skin of the upper arm. Because you may already be in the early exposure window, giving it now, within roughly 4 days of exposure, gives the best chance of preventing smallpox or making it far milder if it does appear. You come back in 28 days for the second 0.5 mL dose to complete the series. Expect a sore arm, maybe some redness and a low-grade fever for a day or two, which is the normal immune response, not the disease. I will give you a sheet with warning signs to watch for in all three of you over the next two to three weeks: a fever above 38.3 C (101 F), severe headache or backache, and especially a rash with firm, deep bumps that all look the same age, which is the pattern that distinguishes smallpox from chickenpox. If any of that appears, you call the number on the sheet immediately and we start tecovirimat. Does that plan feel like something you can both agree to?
{"role":"user"}
Wyatt here. Yeah. Yeah, do it. Start with me so Cody sees it's not a big deal. And... thanks for not talking down to us. Nobody else bothered to actually explain any of this.
{"role":"assistant"}
That means a lot, Mr. Caldwell, and starting with you so Cody sees his dad go first is exactly the right instinct. Let me document the consent for all three of you, note your cardiac history and metoprolol in your chart, and I will draw up the three JYNNEOS doses. You are making a careful, informed decision for your family, and that is all I have ever wanted out of this conversation. Roll up your left sleeve for me.
]
"common_mistakes": [
{"mistake":"Recommending ACAM2000 to a household with cardiac history or eczema without considering JYNNEOS"}— ACAM2000 is a replication-competent vaccinia vaccine carrying risks of myocarditis/pericarditis (~1 in 175 first-time recipients), eczema vaccinatum, and contact transmission to household members. JYNNEOS, a non-replicating MVA vaccine, avoids these and is preferred for patients with cardiac disease, atopic dermatitis, immunocompromise, or whole-household post-exposure vaccination.
(4 more items elided here; the record has 5)
]
"differential_diagnosis": [
{"condition":"Varicella (chickenpox)"}— Lesions appear in successive crops of different ages, are superficial and centripetal (trunk-predominant), and the patient is usually not severely prostrated. Variola lesions are synchronous, deep-seated, firm, umbilicated, and centrifugal with a severe febrile prodrome.
(4 more items elided here; the record has 5)
]
"related_diseases": [
{"disease":"Mpox (monkeypox)"}— A zoonotic orthopoxvirus in the same genus; JYNNEOS and tecovirimat are used for both, and clinical/vaccine experience overlaps substantially.
(4 more items elided here; the record has 5)
]
"executive_summary": A vaccine-hesitant white Oklahoma family presents for post-exposure assessment during a CDC ring-vaccination response after a confirmed deliberate variola (smallpox) release. The parents distrust the Indian-American vaccination officer and cite anti-vaccine media. Through transparent disclosure of real vaccine risks, the clinician distinguishes the replication-competent ACAM2000 vaccine (myocarditis ~1/175, eczema vaccinatum, contact spread) from non-replicating JYNNEOS (0.5 mL SC, two doses 28 days apart), and recommends JYNNEOS for the whole household. A baseline exam catches an irregular pulse and elevated BP, prompting the father's disclosure of concealed metoprolol use for an arrhythmia, which reinforces avoiding ACAM2000. Treatment contingencies include tecovirimat 600 mg PO BID x14 days (400 mg BID weight-based for the 26 kg child) with a CYP3A4-induction caution for midazolam, and second-line brincidofovir/cidofovir with avoidance of concurrent aminoglycosides. Honest, non-condescending counseling converts refusal into informed consent for all three family members.
}
If you see anything wrong please be constructive about it and send in a pull request to fix it. Love you.
Author: Nisten Tahiraj
License: MIT
Similar repos
Contributors
2 commits