Pizov 1999 blast lung injury severity. Mild, moderate, or severe only when inputs match a published definition: mild PaO2/FiO2 > 200, localized infiltrates, no pneumothorax; moderate 60 to 200, diffuse infiltrates, with or without pneumothorax; severe < 60, bilateral infiltrates, and bronchopleural fistula. Unmatched combinations are rejected. Caller assigns all findings. magent does not read a chest radiograph or assay blood gas. Not a ventilator prescription.

Body-mass index from height and weight (Keys/Quetelet). Use for GLP-1 and bariatric prior-auth arithmetic. Deterministic published index with citation; not a diagnosis.

Farris 2017 brain metastasis velocity: bmv = new_brain_metastases / years_since_first_srs, 1 decimal. new_brain_metastases integer 0-200; years_since_first_srs 0.01-30. Low if BMV <4 (median OS 12.4 months after initial distant brain failure), intermediate 4-13 (8.2), high >13 (4.3). Does not floor BMV (NRG-BN009). Not time since last SRS. magent does not count metastases on imaging. Not a diagnosis. Not a medical device.

Celli 2004 BODE index from BMI (≤21 → 1), FEV1 % predicted (0–3), MMRC dyspnea (0–3), and 6-minute walk in meters (0–3). Returns a 0–10 total and quartile 1–4 (0–2 / 3–4 / 5–6 / 7–10). Deterministic published points with citation. Not a GOLD stage and not a transplant listing.

Hesselgard 2007 Behavioural Observational Pain Scale from three caller-assigned 0-2 items (facial_expression, verbalization, body_position) using that paper's 0/1/2 descriptors. magent sums only (max 6) and does not observe the child. No named analgesic cutoff. Not a pain diagnosis or analgesic order. Not FLACC. Not Abbey.

Grossman 2007 Boston Syncope from eight caller-assigned flags: ACS signs/symptoms, conduction disease, worrisome cardiac history, valvular heart disease, family history of sudden death, persistent abnormal ED vitals, volume depletion, and primary CNS event. high_risk (Boston-positive) iff any is true; low_risk iff none. Adult (enrolled 18+). Not a disposition or admission order. magent does not examine the patient or interpret ECG/hematocrit. Not San Francisco Syncope, EGSYS, ROSE, or OESIL.

Bova 2014 seven-point index for confirmed acute PE that is not hypotensive: SBP 90–100 mm Hg +2 (SBP 100 scores; 101 does not; SBP <90 rejected), elevated troponin +2 (caller-assigned lab cutoff; magent does not grade assays), RV dysfunction on echo or CT +2, heart rate ≥110 +1 (110 scores; 109 does not). Max 7. Returns bands 0-2 / 3-4 / >4 (stages I–III) without mortality percents. Not a medical device and not a thrombolysis order.

Manning 1980 fetal biophysical profile. Five caller-assigned variables, each 2 if normal or 0 if abnormal: fetal breathing movements, fetal movements, fetal tone, qualitative amniotic fluid volume, and nonstress test. Score 0-10. 1980 breathing is a 60-second episode in 30 minutes; 1980 fluid is a 1 cm pocket. magent does not ultrasound. Not a delivery order. Does not apply later 30-second breathing, 2 cm pocket, or 8-10/6/<=4 management bands.

Payen 2001 Behavioral Pain Scale from three caller-assigned 1-4 items (facial_expression, upper_limbs, ventilator_compliance) using that paper's Table 1 descriptors. magent sums only (range 3-12) and does not observe the patient. No later pain-present cutoffs. Not a pain diagnosis or analgesic order. Not CPOT (CPOT items are 0-2).

Bergstrom 1987 Braden Scale: six caller-assigned subscales — sensory_perception, moisture, activity, mobility, nutrition (each 1-4) and friction_shear (1-3). magent sums only (range 6-23; lower is higher pressure-sore risk) and returns at_risk true iff score ≤16, the only cutoff in that paper's predictive-validity studies. magent does not examine the patient. Not a medical device, not a diagnosis, and not a treatment plan.

Duca 2020 Brescia-COVID Respiratory Severity Scale testing-criteria count: four caller-assigned binaries (wheezing or inability to speak in full sentences at rest or with minimal effort; respiratory rate >22; PaO2 <65 mmHg or SpO2 <90%; significant worsening on chest radiograph). bcrss is the count of positives (0–4). magent does not examine the patient or read imaging. Not a COVID-19 diagnosis, ventilator order, or medical device. Not the 0–8 NIV/intubation/proning ladder.

Body roundness index from height and waist (Thomas 2013). Models the body as an ellipse: e = sqrt(1 - (waist/(pi*height))^2), BRI = 364.2 - 365.5*e. Deterministic published formula with citation. Not a medical device, VAT measurement, or body-fat diagnosis; magent does not measure the patient.

Lewis 1997 Bristol stool form scale: caller-assigned stool_type 1-7. Returns the assigned type and the published form wording (hard lumps like nuts through watery, entirely liquid). Stool form correlates with whole-gut transit time. magent does not inspect stool. Not a diagnosis, not constipation or diarrhea bands, and not a Rome IBS subtype.

BRUE (Tieder 2016 AAP). brue is true iff the infant is younger than 1 year, the episode was sudden, brief, and now resolved, no explanation remains after appropriate history and physical examination, and at least one of: cyanosis or pallor; absent, decreased, or irregular breathing; marked change in tone; altered level of responsiveness. feature_count is 0-4. Caller assigns seven flags. magent does not examine the infant. Not a diagnosis, not ALTE, not SIDS.

Nama 2022 Supplemental Table 7 logistic probabilities of a serious underlying diagnosis and of event recurrence after a brief resolved unexplained event from age in days, similar-event history, abnormal medical history, prematurity, event clusters, color change, abnormal breathing, and tone change. Returns logits and percents from the unrounded logits. Not a medical device. No admission cutoff. magent does not examine the infant.

Brugada 1991 four-step algorithm for a regular tachycardia with a wide QRS complex. Caller assigns absence of an RS complex in all precordial leads, RS interval >100 ms in one precordial lead, AV dissociation, and VT morphology in V1–V2 and V6. Returns diagnosis vt or svt and the stopping step. magent does not interpret ECGs. Not Brugada syndrome, not Vereckei aVR, not RWPT, not Sgarbossa, and not a cardioversion order. Not a medical device.

Cited Mosteller body surface area. Agents must not invent BSA from model memory; pay per request with x402 USDC on Base. Deterministic published formula with citation. Not a treatment recommendation.

Richards 1990 BUN/creatinine ratio as bun_mg_dl / creatinine_mg_dl, rounded to one decimal. A ratio of 36 or greater suggests upper gastrointestinal bleeding. A ratio below 36 is not helpful in locating the source and is not a lower-GI class. Provide creatinine_mg_dl (0.1-20) or creatinine_umol_l (9-1768), not both. Not a diagnosis, endoscopy order, or Blatchford urea band. magent does not locate a bleed.

Burch 1993 points: °F (<99→0, 99.0–99.9→5, 100.0–100.9→10, 101.0–101.9→15, 102.0–102.9→20, 103.0–103.9→25, ≥104→30); CNS 0/10/20/30; GI-hepatic 0/10/20; HR (<90→0, 90–109→5, 110–119→10, 120–129→15, 130–139→20, ≥140→25); HF 0/5/10/15; AF 10; precipitant 10. Max 140. <25 unlikely, 25–44 impending, ≥45 highly suggestive. temperature_f or temperature_c, not both. Not a diagnosis or ICU order. magent does not measure temperature or an ECG.

Bush 1996 23-item catatonia rating scale from caller-assigned integers. Items 1-16 and 22-23 are 0-3; mitgehen, gegenhalten, ambitendency, grasp_reflex, and perseveration are 0 or 3 only. Severity is the sum of items 1-23 (max 69). The 14-item screen is positive iff at least two of items 1-14 score above 0. magent does not perform the standardized examination. Not a diagnosis, not DSM catatonia criteria, not a medical device.

Bell 2023 C-DIVA from four caller-assigned 0-2 domains (visual appearance, palpable sites, history of difficult access, clinical factors) using that paper's Figure 1 0/1/2 descriptors. magent sums only (max 8) and returns Figure 1 bands low (0-3), medium (4-5), or high (6-8). magent does not examine veins or insert a catheter. Not a vascular-access order, not a medical device.

Ewing 1984 CAGE questionnaire: four caller-assigned yes/no items (cut down, annoyed, guilty, eye-opener), 1 point each, max 4. Screen-positive when two or more answers are affirmative (score ≥2). Deterministic published score with citation; not a diagnosis of alcohol use disorder, not CIWA-Ar, and not AUDIT. magent does not interpret individual items.

Maupain 2016 CAHP nomogram after OHCA. Integer points from age, home vs public arrest, shockable vs non-shockable rhythm, no-flow min, low-flow min, epinephrine tertile (0 / 1-2 / >=3 mg), and arterial pH. Table 3: low <150, medium 150-200, high >200. Figure 2 caption total 177 reads 88% poor outcome. Not a withdrawal-of-care order. magent does not measure pH.

CAM-ICU (Ely 2001). Delirium is present iff Feature 1 (acute onset or fluctuating course) and Feature 2 (inattention) both hold, and Feature 3 (disorganized thinking) or Feature 4 (altered level of consciousness) also holds. The caller assigns each feature as already determined. magent does not observe the patient, administer RASS, or run the attention screening exam. Not a diagnosis and not a medical device.

Griffin 2000 Cambridge logistic risk of undetected type 2 diabetes. p = 1/(1+e^(-λ)); λ = -6.322 - 0.879 if female + 1.222 antihypertensive + 2.191 steroids + 0.063×age + BMI band (<25 0; 25 to <27.5 0.699; 27.5 to <30 1.97; ≥30 2.518) + family (parent or sibling 0.728; parent and sibling 0.753) + smoking (former -0.218; current 0.855). Age 40-64. Probability 0-1. Not FINDRISC, not ADA, not a glucose assay, not a diagnosis.

Canadian C-spine rule (Stiell 2001). Cervical radiography is indicated if any high-risk factor is present (age ≥65, dangerous mechanism, or paresthesias in the extremities). If none, radiography is indicated unless at least one low-risk factor allows safe range-of-motion assessment and the patient can rotate the neck 45° left and right. Not a diagnosis, not NEXUS, and not pediatric. For alert, stable blunt-trauma adults with GCS 15. magent does not examine the neck.

Canadian CT Head Rule (Stiell 2001). CT is indicated if any high-risk factor (GCS <15 at 2 hours, suspected open or depressed skull fracture, any sign of basal skull fracture, vomiting 2 or more times, or age 65 or older) or any medium-risk factor (amnesia before impact >30 min, or dangerous mechanism as assigned by the caller). Not a diagnosis, not NEXUS, not PECARN, not CATCH, and not pediatric. magent does not examine the patient.

Perry 2021 Canadian TIA Score (Table 1) for 7-day stroke or carotid revascularization after TIA. Clinical: first TIA 2, symptoms >=10 min 2, past carotid stenosis 2, antiplatelet 3, gait 1, unilateral weakness 1, vertigo -3, diastolic >=110 3, dysarthria/aphasia 1. Investigations: AF on ECG 2, infarct on CT 1, platelets >=400 2, glucose >=15 mmol/L 3. Age required, 0 points. Total -3 to 23. Bands <=3 low (0.5%), 4-8 medium (2.3%), >=9 high (5.9%). Not ABCD2. Not a diagnosis or admit order.

Robinson 2011 Table 3 pCANRISK (not eCANRISK). Age 0/7/13/15; male 6; BMI 0/4/9/14; waist 0/4/6 by sex; inactivity 1; no daily fruit/veg 2; high BP 4; high glucose 14; macrosomia 1 (women); family first-degree categories ×2; caller-assigned ethnicity White 0 Aboriginal 3 Black 5 East Asian 10 South Asian 11; education 5/1/0. Max 81 F/86 M. Cutoffs 21, 29, 32, 33, 43. Not a diagnosis.

Traube 2014 CAPD: 8 caller-assigned integers 0-4. eye_contact, purposeful_actions, awareness, and communication are reverse-scored in that paper (never contributes 4; 0 is the never-delirium pole after that reverse key). Total 0-32. Screen-positive iff score ≥9. magent does not examine the child or invent a developmental-delay cutoff. Not a psychiatric diagnosis.

Cooperberg 2005 UCSF CAPRA preoperative score (Table 1). Caller-assigned PSA (>=2.1-6 0, >6-10 1, >10-20 2, >20-30 3, >30 4), Gleason 1-3/1-3 0, 1-3/4-5 1, primary 4-5 3, clinical T t1/t2 0 or t3a 1, percent positive biopsies <34 0 or >=34 1, age <50 0 or >=50 1. Score 0-10. Table 4 bands 0_1, 2, 3, 4, 5, 6, ge_7 with 5-year RFS 85/81/66/59/60/34/8. Not Kattan. Not D'Amico. Not CAPRA-S. magent does not assay PSA or read biopsies. Not a diagnosis. Not a medical device.

Caprini 2005 thrombosis risk assessment (Table 3): exclusive age (41-60=1, 61-74=2, >=75=3) and exclusive surgery (minor=1; arthroscopic/major/laparoscopic=2; arthroplasty=5) plus caller-assigned 1-, 2-, 3-, and 5-point flags. Max 78. Bands: 0-1 low, 2 moderate, 3-4 high, >=5 highest. Not a diagnosis. Not Caprini 2009/2013. Not a prophylaxis order.

VanWagner 2017 CAR-OLT integer points (Table 3) for 1-year CVD complication after first orthotopic liver transplant: age bands, sex, caller-assigned race (white 7, black 10, other 0), work, education, atrial fibrillation 25, ventilator 13, pulmonary hypertension 9, no HCC 6, hypertension 4, diabetes 4, heart failure 7. Table 4 predicted percent. Not a listing decision. magent does not assign race.

Cardiac Anesthesia Risk Evaluation (CARE) score (Dupuis 2001). Caller-assigned class 1-5; emergency suffix E only for classes 3-5 (labels 3E, 4E, 5E). Returns care_label plus Dupuis 2001 Table 4 in-hospital mortality and morbidity percents from the Ottawa reference population. Not an anesthetic plan or surgical booking. magent does not examine the patient. Not Parsonnet, EuroSCORE, or Gupta MICA.

Hurria 2011 Table 5 CARG toxicity: age ≥72 (+2), GI/GU (+2, not inverted), standard dose (+2), polychemotherapy (+2), Hb male <11 or female <10 g/dL (+3), caller-assigned Jelliffe CrCl <34 (+3; magent does not compute Jelliffe), hearing fair/worse (+2), ≥1 fall in 6 months (+3), limited walking 1 block (+2), help with meds (+1), decreased social activity (+1). Max 19. 0-5 low 30%, 6-9 intermediate 52%, 10-19 high 83%. Not the 2016 revision, not CRASH, not a chemo order.

CAROC 2010 10-year MOF category from age, sex, femoral-neck T-score, and fragility or glucocorticoid flags. Basal low/moderate/high from printed T-score bands; hip or spine fragility or multiple fractures is high; fracture after 40 plus steroids is high; either flag alone bumps one class. Labels <10%, 10-20%, >20%. Not FRAX or a drug order. Not a medical device.

CART integer score from Churpek 2012 Table 3 (beta x 9): maximum respiratory rate, maximum heart rate, minimum diastolic BP, and age, summed to 0-57. magent does not measure vitals; the caller assigns the numbers. Deterministic published arithmetic with citation; not a cardiac-arrest diagnosis and not an RRT order. Does not return risk-band percents.

Lee 2020 Coronavirus Anxiety Scale from five caller-assigned integers 0-4 over the past two weeks (dizzy, sleep, paralyzed, appetite, nausea). magent does not interview the patient. 0 not at all, 1 rare less than a day or two, 2 several days, 3 more than 7 days, 4 nearly every day. Total 0-20. Returns screen_ge_9 (score ≥9; this paper 90% sensitivity and 85% specificity). Not a diagnosis, not GAD-7, and not extra mild/moderate bands.

Taylor 2006 CASPAR classification criteria for psoriatic arthritis. Required: inflammatory_articular_disease true (joint, spine, or enthesis). Score: current psoriasis 2; personal history 1 only if current is false; family history 1 only if current and personal are false; nail dystrophy 1; RF-negative 1; dactylitis (current or history) 1; juxtaarticular new bone 1. Classified iff score ≥3 (max 8). Not a diagnosis. Not Moll and Wright. magent does not examine skin or read films.

Jones 2009 COPD Assessment Test (CAT): eight caller-assigned integers 0-5 (cough, phlegm, chest tightness, breathlessness, activities, confidence, sleep, energy). Sum 0-40. magent does not interview. Jones 2009 printed no GOLD 0-10/11-20/21-30/31-40 impact bands; later website cutoffs are not applied. Not a diagnosis, not GOLD ABE, not BODE, and not a medical device.

CATCH (Osmond 2010). CT indicated if any of 7 caller-assigned findings after pediatric minor head injury. High (neurologic intervention): GCS <15 at 2h, open/depressed skull fracture, worsening headache, irritability. Medium if not high (brain injury on CT): basal skull fracture, large boggy scalp hematoma, dangerous mechanism. Low if none. Not PECARN, CHALICE, Canadian CT Head, NEXUS, or CATCH-2. magent does not examine the child or read CT. Not a diagnosis or medical device.

Corrected count increment (CCI) after platelet transfusion: (post - pre platelet count per uL) * BSA_m2 / (dose in 10^11). Davis 1999 Transfusion. Returns CCI and the uncorrected increment only. Caller supplies BSA; magent does not compute Nadler BSA from height/weight, assay platelets, emit a 5000 or 7500 refractoriness cutoff, or order a transfusion. Davis 1999 cautions CCI/PPR can distort comparisons of platelet products.

CCS angina of effort class 1-4 as assigned by the caller (Campeau 1976 Table 1). Returns class, score, max 4, and the published class description. Class 1-4 only; no class 0. magent does not walk the patient, interpret a treadmill, or examine the patient. Not an ACS diagnosis, not NYHA, and not a silent-ischemia screen.

Best 1976 Crohn's disease activity index (CDAI): eight weighted variables (7-day liquid stools *2, abdominal-pain sum *5, well-being sum *7, extra-intestinal count *20, antidiarrheal *30, abdominal mass 0/2/5 *10, hematocrit deficit *6, caller-assigned weight-percent deviation). magent returns the rounded integer index only and does not apply later trial activity bands. magent does not examine the patient. Not a Crohn's diagnosis and not the Harvey-Bradshaw one-day index.

Aletaha 2005 Clinical Disease Activity Index (CDAI) for rheumatoid arthritis: SJC28 + TJC28 + patient global (0–10 cm) + evaluator global (0–10 cm). No CRP. Half points allowed; the sum is not integer-rounded. magent does not count joints. Arthritis Res Ther 2005 (ar1740) validates the sum and does not publish activity-state cutoffs. Not SDAI. Not Crohn's CDAI. Not a rheumatoid arthritis diagnosis.

ACIP 2025 Table 2 catch-up minimum age and previous-dose interval. Eligible if age_months meets the printed minimum and weeks_since_previous_dose meets a printed interval (omit for dose 1). HepB dose 2 is 4 weeks. MMR/HepA/varicella dose 1 is 12 months. Not a vaccine order. magent does not immunize. Not routine age bars, Hib/PCV branching, HepB dose 3, or the 4-day grace period.

Weiss 2014 Community ED Overcrowding Scale (CEDOCS) reduced model: critical-care patients, longest admitted-patient hours since admission, waiting-room count, ED patients indexed to ED beds, and annual ED visits with a restricted cubic spline. Returns the 1-decimal score on the paper's 0-100 scale without clamping or named class bands. Caller supplies counts and hours; magent does not count patients or wait times. Not a medical device and not a diversion or ambulance-bypass order.

Centor or McIsaac (age-modified) score for streptococcal pharyngitis from four caller-assigned findings. edition=centor (max 4) or edition=mcisaac (age 3-14 +1, 15-44 0, >=45 -1, max 5) is required; magent does not pick one. Deterministic published score with citation; not a test-or-treat protocol.

Rockwood 2005 CSHA Clinical Frailty Scale 1-7 as assigned by the caller (Box 1). Returns cfs, score, max 7, and the published label. Does not implement the later 9-point CFS (8 very severely frail, 9 terminally ill) or CFS 2.0 label changes. Not a diagnosis; magent does not examine the patient.

Liang 2020 COVID-GRAM logistic probability of critical illness (ICU, invasive ventilation, or death) among hospitalized COVID-19 from age, caller-assigned chest radiograph abnormality, hemoptysis, dyspnea, unconsciousness, comorbidity count, cancer history, neutrophil-to-lymphocyte ratio, LDH (U/L), and direct bilirubin (umol/L). Table 3 coefficients. magent does not read the film or count comorbidities. Not a diagnosis or ICU-allocation order.