Grobman 2007 first prenatal visit logistic probability of vaginal birth after cesarean from maternal age, prepregnancy BMI, caller-assigned ethnicity (african_american, hispanic, or other), any prior vaginal delivery, prior VBAC after cesarean, and recurrent cesarean indication. Ethnicity uses the paper terms; magent does not assign race or ethnicity. Not the 2009/2010 admission model and not Grobman 2021. Not a trial-of-labor order.

Grogan 2016 ATTRwt staging: 1 point when NT-proBNP >3000 ng/L; 1 point when cardiac TnT >0.05 ng/mL. Stage I = 0, II = 1, III = 2. Equal to a cutoff does not score. Cohort 4-year OS I 57% / II 42% / III 18%; stage III median 20 months (labels, not a prediction). Troponin T only. Not Gillmore NAC (no eGFR). Not Mayo AL. Not NYHA. magent does not assay the biomarkers.

Gupta 2011 Table 2 logistic probability of 30-day myocardial infarction or cardiac arrest after surgery from age, ASA class, functional status, creatinine category (normal, abnormal >1.5 mg/dL, or missing), and surgery type (hernia reference). Returns logit and estimated_probability_percent. Not a medical device, not a diagnosis, not surgical clearance, and not RCRI. Gupta 2011 publishes no low/high cutoff.

Gupta 2013 Mayo Clin Proc logistic probability of postoperative pneumonia from age, COPD, ASA class, functional status, preoperative sepsis, smoking in the last year, and surgery type (hernia reference). Returns logit and estimated_probability_percent. Not a medical device, not a diagnosis, not an antibiotic order, and not a ventilator order. Gupta 2013 publishes no low/high cutoff and does not use race.

Gupta 2011 Chest logistic probability of postoperative respiratory failure (mechanical ventilation >48 h after surgery or unplanned intubation within 30 days) from ASA class, functional status, preoperative sepsis, emergency case, and surgery type (hernia reference). Returns logit and estimated_probability_percent. Not a medical device, not a diagnosis, not a ventilator order, and not ARISCAT. Gupta 2011 publishes no low/high cutoff.

Peterson 2010 GWTG-HF Figure 2 integer points for in-hospital mortality in hospitalized heart failure: age 0-28, systolic BP 0-28, BUN 0-28, sodium 0-4, heart rate 0-8, COPD +2, caller-assigned nonblack (black=false) +3. Total 0-100. magent does not assign race. Not a diagnosis.

Reddy 2018 H2FPEF points score. Six components, max 9: BMI >30 (2), >=2 antihypertensives (1), atrial fibrillation (3), echo PASP >35 mm Hg (1), age >60 (1), E/e' >9 (1). BMI 30 and age 60 do not score. Returns score and max_score only; odds roughly double per point. Not probability classes, not the continuous nomogram, not HFA-PEFF, not a diagnosis or cath substitute. magent does not read echo or compute BMI.

Duan 2017 HACOR (ESM Table 2): HR ≤120 → 0, ≥121 → 1; pH ≥7.35 → 0, 7.30–7.34 → 2, 7.25–7.29 → 3, <7.25 → 4; GCS 15 → 0, 13–14 → 2, 11–12 → 5, ≤10 → 10; P/F ≥201 → 0, 176–200 → 2, 151–175 → 3, 126–150 → 4, 101–125 → 5, ≤100 → 6; RR ≤30 → 0, 31–35 → 1, 36–40 → 2, 41–45 → 3, ≥46 → 4. Max 25. Low ≤5, high >5. Not an intubation or NIV-failure order. magent does not read an ABG, GCS the patient, or a ventilator.

Hamilton 1959 HAM-A: 14 caller-assigned integers 0-4 (0 not present, 1 mild, 2 moderate, 3 severe, 4 very severe) for anxious mood, tension, fears, insomnia, intellectual, depressed mood, somatic muscular, somatic sensory, cardiovascular, respiratory, gastrointestinal, genitourinary, autonomic, and behaviour at interview. Sum 0-56. magent does not interview the patient. No total-score severity bands in Hamilton 1959; later 17/24 cutoffs are not applied. Not a diagnosis and not a medical device.

Hamilton 1960 17-variable depression rating (not the later 21-item form). Eight caller-assigned 0-4 items and nine 0-2 items (agitation is three-point in that paper). Sum 0-50. magent does not interview. Hamilton 1960 printed no total-score severity bands; later 0-7/8-16/17-23/>=24 cutoffs are not applied. Not a diagnosis and not a medical device.

HAPS (Lankisch 2009). Harmless iff rebound tenderness and/or guarding is absent, hematocrit is normal, and serum creatinine is normal. The caller assigns each flag as already determined. magent does not examine the abdomen and does not measure hematocrit or creatinine. No millimetre or sex lab cutoffs. Not a medical device, not a diagnosis, and not a discharge order.

Sohal 2007 HARK: four caller-assigned yes/no flags (humiliation, afraid, rape, kick), 1 point each, max 4. Screen-positive when score ≥1 (ROC-optimal cut-off in Sohal 2007). Deterministic published score with citation; not a diagnosis of intimate partner violence and not a medical device. magent does not interview the patient. Not HITS 1-5 scoring. Not the AAS pregnancy item (HARK removed it).

Harvey and Bradshaw 1980 simple Crohn's-disease activity index (one-day; items 1-3 previous day). Well-being 0-4, abdominal pain 0-3, liquid stools (1 point each; magent clamp 0-30), abdominal mass 0-3, plus 1 per listed complication (arthralgia, uveitis, erythema nodosum, aphthous ulcers, pyoderma gangrenosum, anal fissure, new fistula, abscess). magent sums only (max 48) and does not return later remission cutoffs. Not a Crohn's diagnosis or Best 1976 CDAI. magent does not examine the patient.

HAS-BLED score from discrete clinical flags. Use for bleeding-risk arithmetic in atrial fibrillation. Deterministic published score with citation.

HAT (Lou 2008 Neurology) for sICH after stroke thrombolysis. Diabetes or glucose >200 mg/dL (1; not additive). NIHSS <15 0, 15-20 1, >20 2. CT hypodensity none 0, <1/3 MCA 1, >1/3 MCA 2 (no equals-1/3). Max 5. Collective sICH 2/5/10/15/44% for scores 0/1/2/3/>3. Glucose in mg/dL (mmol × 18). Not a tPA withhold. magent does not assign NIHSS or read CT.

Citterio 2016 JAMA Surgery hierarchic classes for liver decompensation after HCC resection from caller-assigned portal_hypertension, major_hepatectomy, and meld (integer 6-40). High if PHT and major (LD 60.0%, liver-related mortality 25.0%). Low if no PHT, minor, and MELD ≤ 9 (LD 4.9%, mortality 4.4%). Else intermediate (LD 28.6%, mortality 9.0%). magent does not compute MELD or HVPG. Not a surgical indication, transplant listing, or medical device.

O'Mahony 2014 HCM Risk-SCD: 5-year SCD probability from wall thickness, LA diameter, LVOT gradient, family history of SCD, NSVT, unexplained syncope, and age. Returns prognostic_index, five_year_scd_risk_percent, and 4%/6% risk_band (lower / intermediate / higher). Not an ICD or device order. Not a medical device.

Sorror 2005 HCT-specific comorbidity index. Seventeen caller-assigned comorbidities with published weights 1, 2, or 3 (max 26). Pulmonary moderate vs severe and hepatic mild vs moderate/severe are mutually exclusive. Risk groups: 0 low, 1–2 intermediate, ≥3 high. Caller supplies the published category; magent does not interpret PFTs or echo and does not list for transplant. Not Charlson. Not a diagnosis. Not a medical device.

Alghamdi 2019 HE-MACS ACS probability from age, sex, caller-assigned observed sweating, acute ECG ischaemia, right-arm or shoulder radiation, vomiting, hypotension (SBP <100), and current smoking. Table 2 printed coefficients. Returns probability only; four-group ROC thresholds are in that paper's supplement and are not restated here. magent does not interpret ECG or take a history. Not MACS (Body 2014 biomarkers) or T-MACS. Not a diagnosis.

Cappelli 2012 HEADS-ED from seven caller-assigned 0-2 domains (Home, Education, Activities/peers, Drugs/alcohol, Suicidality, Emotions/behavior, Discharge resources). 0 no action needed, 1 needs action but not immediately, 2 needs immediate action. magent sums only (max 14) and does not interview a child. No labeled high-risk band. Not a medical device or psychiatric admission order.

HEAR score for emergency chest pain from age and caller-assigned history, ECG, and risk factors (each 0-2, max 8). No troponin. Smith 2020 very-low-risk group is scores 0 and 1 (MACE ~0.9% in that paper). Deterministic published score with citation; not a diagnosis or discharge protocol. magent does not interpret history, count risk factors, or read an ECG.

HEART score for emergency chest pain from age and caller-assigned history, ECG, risk factors, and troponin (each 0-2, max 10). Bands 0-3, 4-6, and 7-10 from Six 2008. Deterministic published score with citation; not a diagnosis or discharge protocol. magent does not read an ECG.

Mahler 2015 HEART Pathway for emergency chest pain. early_discharge_candidate is true only when the caller-assigned HEART score is 0-3 and both 0-hour and 3-hour troponin flags are not elevated (above the 99th percentile). HEART items are not assigned here. magent does not read an ECG or assay a troponin. Does not order a stress test or angiogram. High-risk features do not mandate a test. Not a medical device and not a diagnosis.

Davis 2017 HEAVEN criteria from six caller-assigned flags: hypoxemia (SpO2 ≤93% at initial laryngoscopy as assigned), extremes of size (pediatric ≤8 years or clinical obesity as assigned), anatomic challenges, vomit/blood/fluid, exsanguination, and limited neck mobility. high_risk iff any is true; criteria_met is the count of true flags (0-6); low_risk iff none. Not an intubation order. magent does not measure SpO2 or examine the airway. Not Mallampati, not EGRI.

Gage 2006 HEMORR2HAGES (NRAF) AF major-bleed score: prior bleed 2; hepatic or renal, ethanol, malignancy, age >75, reduced platelet count or function, uncontrolled hypertension, anemia, genetic factors, excessive fall risk, and stroke each 1. Max 12. Warfarin major bleed per 100 patient-years: 0→1.9, 1→2.5, 2→5.3, 3→8.4, 4→10.4, ≥5→12.3. Not a medical device. Not a decision to start or stop anticoagulation. Not HAS-BLED, ORBIT, or ATRIA.

Cuker 2010 HEP score: eight features with signed weights (range -16 to 19). Magnitude of platelet fall, typical- or rapid-onset timing, nadir, thrombosis (at most one), SQ heparin skin necrosis, IV-bolus systemic reaction, bleeding, and other causes (select all; no other cause +3). Cutoff 2 was 100% sensitive / 60% specific vs the expert panel. Not a HIT diagnosis or 4Ts score. magent does not assay SRA or ELISA.

Rodger 2008 HERDOO2 after a first unprovoked VTE. Score 1 each for hyperpigmentation, edema, or redness in either leg; D-dimer ≥250 μg/L while on anticoagulants; BMI ≥30; age ≥65 (equality scores). Max 4. low_risk is true only for women with score ≤1. Men have no low-risk subgroup in the derivation. Not a medical device and not a duration-of-anticoagulation order.

Hestia criteria (Zondag 2011). Outpatient PE is eligible only when none of 11 caller-assigned exclusions hold: hemodynamic instability, thrombolysis or embolectomy, high bleeding risk, oxygen >24 h, PE on anticoagulation, IV pain medication >24 h, medical or social admission, CrCl <30, severe liver impairment, pregnancy, or documented HIT. Any yes excludes home treatment. Not a diagnosis, not PESI, not sPESI.

Lyon 2020 HFA-ICOS baseline cardiovascular risk before anthracycline chemotherapy. Returns risk_level low, medium, high, or very_high from published Very high, High, Medium 2, and Medium 1 factors, plus medium_points and the paper's expert-opinion cardiotoxicity_guide. Age and LVEF use the highest matching printed band only. Caller assigns labs and BMI >30. magent does not examine the patient or read imaging. Not a medical device or treatment order.

Lyon 2020 HFA-ICOS baseline cardiovascular risk before second- or third-generation multi-targeted kinase inhibitors for CML targeting BCR-ABL. Class is very_high, high, medium, or low from published very-high and high factors plus Medium 1/2 points (0–1 low, 2–4 medium, ≥5 high). Caller assigns the 10-year CVD >20% threshold; magent does not compute that score. Not a medical device or treatment order.

Lyon 2020 HFA-ICOS baseline cardiovascular risk before HER2-targeted therapy (trastuzumab and related agents). Very high, high, medium, or low from published factors plus Medium 1/2 points; expert-opinion cardiotoxicity guide <2% / 2-9% / 10-19% / >=20%. Concurrent anthracycline is High; sequential anthracycline before HER2 is Medium 1. Deterministic published classification with citation; not a treatment order. magent does not measure LVEF or biomarkers.

Lyon 2020 HFA-ICOS baseline CV risk before proteasome inhibitors and immunomodulatory drugs for multiple myeloma. Returns risk_level low, medium, high, or very_high plus medium_points and the paper's expert-opinion cardiotoxicity_guide. Age >=75 is High; 65-74 is Medium 1. Caller assigns labs, BMI >30, and high-dose dexamethasone. magent does not examine the patient or read imaging. Not a medical device or treatment order.

Lyon 2020 HFA-ICOS baseline CV risk before combination RAF and MEK inhibitors. Very high if heart failure or cardiomyopathy; High if MI/CABG, stable angina, severe valvular disease, prior anthracycline, or ≥5 medium points; Medium if 2-4 medium points; Low if 0-1. Age ≥65 is Medium 1 only. LVEF 50-54 is Medium 2; this table has no printed LVEF <50 High row. Expert-opinion cardiotoxicity guide <2% / 2-9% / 10-19% / ≥20%. Not a treatment order.

Lyon 2020 HFA-ICOS baseline CV risk before VEGF inhibitors or VEGF TKIs. Very high if heart failure/cardiomyopathy or arterial vascular disease. High if venous thrombosis, LVEF <50%, QTc >=480 ms, age >=75, hypertension, prior anthracycline, or >=5 medium points. Medium if 2-4 medium points. Guide <2% / 2-9% / 10-19% / >=20%. Not a device or treatment order.

High-dose insulin euglycemia therapy (HIET) from Engebretsen 2011: 1 U/kg insulin bolus, then a 1-10 U/kg/h continuous infusion. Weight in kg or lb. Not a prescription. Used in beta-blocker or calcium-channel-blocker cardiogenic shock. magent does not titrate, choose dextrose, or add DigiFab, glucagon, or calcium. Deterministic published arithmetic with citation.

Hack 2014 H-Impairment Index (HII). Five caller-assigned bedside tasks (gross motor, mentation and speech, tracing curve, nystagmus, finger-to-nose), each integer 0-4. HII is completed-sum / (4 × completed tasks); denominator 20 if all five; range 0-1. Omit a task that was not attempted. Serial decline is not modeled. Does not take blood alcohol concentration. Not a diagnosis, discharge order, or legal sobriety test. magent does not examine the patient.

Kattah 2009 three-step HINTS (Head-Impulse, Nystagmus, Test-of-Skew) for patients already in acute vestibular syndrome (AVS). Three caller-assigned flags: head_impulse_normal, direction_changing_nystagmus, and skew_deviation. dangerous is true if any flag is true (INFARCT: Impulse Normal, Fast-phase Alternating, Refixation on Cover Test). benign iff none. Not HINTS-plus (hearing). Not a stroke diagnosis or a substitute for neuroimaging. magent does not examine the patient.

Smith 2012 seven-item HIV incidence risk index for MSM. Age 18-28=8, 29-40=5, 41-48=2, >=49=0; male partners 0-5=0, 6-10=4, >=11=7; HIV-positive partners 0=0, 1=4, >=2=8; unprotected receptive anal 0=0, >=1=10; insertive anal with HIV-positive partner 0-4=0, >=5=6; poppers 3; amphetamines 5. Range 0-47; at_or_above_10 when score >=10. Caller-assigned. Not a diagnosis or PrEP order.

Sherin 1998 HITS from four caller-assigned integers 1-5 (hurt, insult, threaten, scream; never=1 to frequently=5). magent does not interview the patient or interpret the items. Sum 4-20. Screen-positive when score ≥10. Deterministic published score with citation; not a diagnosis of intimate partner violence and not a medical device.

Ho 2004 day-3 IV-steroid score for severe UC (0–9): mean stools (<4 → 0; 4 ≤ x < 6 → 1; 6 ≤ x ≤ 9 → 2; x > 9 → 4), caller-assigned colonic dilatation (+4), albumin <30 g/L (+1; 30 is not low). Failure: 0–1 → 11%, 2–3 → 43%, ≥4 → 85% (ROC ≥4). Not Travis, not Truelove and Witts, not a colectomy order. magent does not count stools, read films, or measure albumin.

HOMA-IR as (fasting glucose in mmol/L × insulin in µU/mL) / 22.5 (Matthews et al. 1985). Provide glucose_mg_dl (50–1000) or glucose_mmol_l (2.8–55.5), not both; mg/dL converts as mmol/L = mg/dL / 18. Returns the approximation rounded to two decimals. Not HOMA-B, not the HOMA2 computer model, not a diabetes diagnosis, and not an insulin-resistance cutoff.

Pasquier 2018 HOPE survival probability after ECLS for hypothermic cardiac arrest from sex (published male coefficient, not race), asphyxia, age, potassium, CPR duration, and temperature. Returns logit, estimated_survival_percent from the unrounded logit, and Table 2 HOPE >= 10% flag. Not an ECLS or rewarming order. magent does not cannulate.

HOSPITAL (Donze 2013 JAMA Intern Med acronym table): hemoglobin_lt_12 (last Hb <12 g/dL) 1; oncology_service 2; sodium_lt_135 (last Na <135 mmol/L) 1; procedure (any ICD-coded procedure during the stay) 1; urgent_or_emergent_admission 1; admissions_12mo 0-1→0, 2-5→2, >5→5; los_ge_5_days 2. Max 13. Bands 0-4 low, 5-6 intermediate, >=7 high for potentially avoidable 30-day readmission (SQLape), not all-cause. Flags and counts are caller-assigned. Not a discharge order or diagnosis.

Fardet 2014 HScore (Table 4; max 337) from nine caller-assigned items: immunosuppression 18, temperature 0/33/49, organomegaly 0/23/38, cytopenia lineages 0/24/34 (1/2/3), ferritin 0/35/50, triglycerides 0/44/64, fibrinogen 0/30, SGOT 0/19, marrow hemophagocytosis 35. Best cutoff 169. magent does not take raw labs; all nine required. Not HLH-2004. Not a diagnosis. Not a medical device.

Hunt and Hess 1968 SAH surgical-risk grade from a caller-assigned grade 1-5. Serious systemic disease (as assigned by the caller) places the patient in the next less favorable category, capped at 5. Deterministic published grade with citation. Not a diagnosis; not WFNS; not Hunt-Kosnik 1974 grade 0/1a.

Karter 2017 Figure 2 T2D 12-month hypoglycemia-related ED/hospital-use risk from prior_hypo_ed_hospital_count (0-20), ed_visits_12mo (0-50), insulin, sulfonylurea, severe_or_esrd (dialysis or CKD 4-5, eGFR <=29), and age_lt_77. First-match 6 options: >=3 prior or 1-2 plus insulin = high (>5%); remaining rows intermediate (1-5%) or low (<1%). Not a points sum. magent does not assay glucose. Not a diagnosis. Not an insulin dose.

Dellon 2022 I-SEE Table I: symptoms 0/1/2/4, food_impaction 0/2/4, stacking complications (hospitalization 4, perforation 15, malnutrition 15, persistent_inflammation 15), inflam endoscopy 0-2, peak eos 0-2, fibro endoscopy 0/1/2/15, histologic fibrosis 2. Max 78. Bands <1 inactive, 1-6 mild, 7-14 moderate, ≥15 severe. magent does not perform endoscopy or count eos. Not EREFS. Not a diagnosis.

Devine ideal body weight and 0.4 adjusted body weight from height, sex, and actual weight. Deterministic published formula with citation; not a dosing recommendation.

ICC 2022 prefibrotic and overt PMF (Arber 2022). Diagnosed iff 3 majors and ≥1 minor. Majors: megakaryocyte morphology, not another myeloid neoplasm, JAK2/CALR/MPL or other clone or no reactive fibrosis. Minors: anemia not from comorbidity, WBC ≥11e9/L, splenomegaly, LDH above ULN; leukoerythroblastosis counts only in overt PMF. Not DIPSS, MIPSS70, WHO 5th, or post-PV/post-ET MF. magent does not read fibrosis grade or sequence drivers. Not a diagnosis or medical device.

ICC 2022 systemic mastocytosis (Arber 2022). Met when major infiltrates (≥15 tryptase- and/or CD117-positive MCs in aggregates in BM or extracutaneous organ) alone, or ≥3 minors. Minors: >25% spindle or atypical immature morphology, KIT D816V or other activating KIT mutation, CD25/CD2/CD30, tryptase >20 ng/mL (not counted in SM-AMN). Not a diagnosis. Not Valent 2016 WHO SM. Not WHO 5th/Khoury. Not cutaneous mastocytosis only. magent does not read a biopsy or assay tryptase.