End Hazing in Medicine

Who we are and where every number comes from.

Evidence first. Every claim traces to a source we read.

About

End Hazing in Medicine is a public advocacy project. Our goal: training schedules that sustain health, hourly pay that reflects skill, and debt policy that rewards service.

We cite every stat to a source we read. We correct errors promptly. We never invent testimony. We give no medical advice. If you are in crisis, call or text 988.

Site illustrations are AI-generated; no photograph on this site depicts a real resident.

Close-up of a resident's hands holding a clipboard with patient charts

How to read our numbers

Each claim on this site has one row in our stat table. Each row names the primary source and its URL. Where we found a second source, we list it. Where we could not, we mark GAP. Math we did ourselves shows inputs plus steps.

Key site arithmetic, all reproducible: $68,166 / 52 = $1,310.88/week. At 80 hours: $16.39/hr[7]. Surgery (83.1 hrs): $15.77/hr[5]. Psychiatry (49.5 hrs): $26.48/hr[6]. Overtime value at 80 hours: $170,414/year vs. $68,166 stipend — a $102,248 gap[34].

Every claim traced: stat-table row count 104 stat-table rows — every numeric claim traced to a page read this run Counted from content/stat-table.md at build time.

Sources — every stat-table row

Each entry below is one row of our stat table, in row order. Superscript numbers in our articles (like [7]) match these row numbers exactly.

Pay, hours & overtime math rows 1–11

  1. PGY-1 avg stipend, 2025 AAMC survey (350 institutions, 114,361 residents) — $68,166. AMA, "Resident physician pay still rising, but growth trails inflation" (read 2026-09-29.) Second source: MedSchoolCoach, "Average Residency Salary by PGY".
  2. Institutions paying same base stipend across all specialties — 96.9%. AMA, "Resident physician pay still rising, but growth trails inflation" (same source as #1) (read 2026-09-29.) Second source: Resident Insider, "Resident Pay".
  3. General surgery self-reported hours (Panacea 2025 survey, n=479) — 83.1 hr/wk. Resident Insider, "Resident Pay" (read 2026-09-29.) Second source: GAP (Panacea report itself not opened; survey n=479, ±3.8%, Jan–Feb 2025 per page).
  4. Psychiatry self-reported hours, same survey — 49.5 hr/wk. Resident Insider, "Resident Pay" (same source as #3) (read 2026-09-29.) Second source: GAP.
  5. Surgery effective hourly: $68,166/52/83.1 — $15.77/hr. Site calculation from #1 + #3. (read 2026-09-29.)
  6. Psych effective hourly: $68,166/52/49.5 — $26.48/hr. Site calculation from #1 + #4. (read 2026-09-29.)
  7. 80-hr-cap hourly: $68,166/52/80 — $16.39/hr. Site calculation from #1 + #8. (read 2026-09-29.)
  8. ACGME max clinical/educational work, incl. home work + moonlighting — 80 hr/wk avg 4 wks. ACGME Common Program Requirements (Residency), eff. 7/1/2026, §6.20 (read 2026-09-29.) Second source: KFF Health News citing 2021 CPR.
  9. Max in-house call + handoff extension — 24 hr + up to 4 hr. ACGME Common Program Requirements (Residency), eff. 7/1/2026, §6.20 (same source as #8) (read 2026-09-29.) Second source: KFF Health News citing 2021 CPR.
  10. Free time required after 24-hr call — 14 hr. ACGME Common Program Requirements (Residency), eff. 7/1/2026, §6.20 (same source as #8) (read 2026-09-29.) Second source: GAP.
  11. Minimum free days — 1 day in 7, avg 4 wks. ACGME Common Program Requirements (Residency), eff. 7/1/2026, §6.20 (same source as #8) (read 2026-09-29.) Second source: GAP.

Debt, tax & labor law rows 12–21

  1. Avg total education debt, class of 2025 (incl. premed loans) — $223,130. Credible, "Average Medical School Debt" (citing AAMC) (read 2026-09-29.) Second source: AMA cites AAMC avg $212,341 (earlier cohort).
  2. Median med-school debt $200,000; median premed debt $28,000 — $200k / $28k. Credible, "Average Medical School Debt" (citing AAMC) (same source as #12) (read 2026-09-29.) Second source: GAP.
  3. Graduates leaving school with debt — 70%. Credible, "Average Medical School Debt" (citing AAMC) (same source as #12) (read 2026-09-29.) Second source: AMA cites AAMC avg $212,341 (earlier cohort).
  4. Graduates with >$300k total debt — 28% (~1 in 4). Credible, "Average Medical School Debt" (citing AAMC) (same source as #12) (read 2026-09-29.) Second source: GAP.
  5. 4-yr cost of attendance: public / private — $286k+ / $390k+ (2024); $297,745 / $408,150 (2026). AMA cites AAMC avg $212,341 (earlier cohort) (same source as #12) (read 2026-09-29.) Second source: Credible, "Average Medical School Debt" (citing AAMC).
  6. FLSA overtime rule for covered workers — 1.5x past 40 hr/wk. Cornell LII, 29 USC §207 (read 2026-09-29.) Second source: GAP.
  7. FLSA exemption for bona fide executive/admin/professional employees — exemption exists. Cornell LII, 29 USC §213(a)(1) (read 2026-09-29.) Second source: GAP (whether courts apply it to residents is a legal reading, not asserted here).
  8. Federal minimum wage — $7.25/hr. Cornell LII, 29 USC §206 (read 2026-09-29.) Second source: GAP.
  9. Student-loan interest deduction cap (lesser of cap or interest paid; MAGI phaseout) — $2,500/yr. IRS Topic 456 (read 2026-09-29.) Second source: GAP.
  10. Indebted grads seeking forgiveness who planned PSLF — ~90%. AMA cites AAMC avg $212,341 (earlier cohort) (same source as #12) (read 2026-09-29.) Second source: GAP.

Burnout, depression & crisis help rows 22–27

  1. 2nd-yr residents with ≥1 burnout symptom (n=3,588, JAMA Dyrbye 2018) — 45.2%. CMAJ summary of JAMA study (read 2026-09-29.) Second source: 2 Minute Medicine summary ("nearly half").
  2. Career-choice regret, same study — 14.1%. CMAJ summary of JAMA study (same source as #22) (read 2026-09-29.) Second source: 2 Minute Medicine summary ("nearly half").
  3. Residents with depression/depressive sx (JAMA Mata 2015: 54 studies, 17,000+ residents) — ~1/3; range 20.9–43.2%. AMA, "Depression in medical residents" (read 2026-09-29.) Second source: AAO EyeNet (28% figure).
  4. Major depressive episode in training vs same-age public — 28% vs 8%. AAO EyeNet (28% figure) (same source as #24) (read 2026-09-29.) Second source: AMA, "Depression in medical residents".
  5. Interns with suicidal ideation ≥once during intern year — 16%. AAO EyeNet (28% figure) (same source as #24) (read 2026-09-29.) Second source: GAP.
  6. 988 Lifeline: call/text/chat 988; 24/7; free + confidential — 988. 988 Suicide & Crisis Lifeline official site (read 2026-09-29.) Second source: SAMHSA 988 page ("24/7, free and confidential").

On-record voices & context rows 28–38

  1. González: county residents earn "less than $18 an hour" (his estimate) — <$18/hr. KFF Health News, May 2022 (read 2026-09-29.) Second source: GAP (our math #7 corroborates magnitude).
  2. Most LA County residents report 80-hr weeks (CIR member survey via Naranjo) — "most" @ 80 hr. KFF Health News, May 2022 (same source as #28) (read 2026-09-29.) Second source: GAP.
  3. CIR members / share of US trainees — 20,000+ / ~1 in 7. KFF Health News, May 2022 (same source as #28) (read 2026-09-29.) Second source: GAP.
  4. Avg resident salary 2021 (Medscape via Becker's) — $64,000. KFF Health News, May 2022 (same source as #28) (read 2026-09-29.) Second source: GAP.
  5. Residents feeling adequately compensated (Medscape 2021) — 43%. KFF Health News, May 2022 (same source as #28) (read 2026-09-29.) Second source: GAP.
  6. Training length before independent practice — 3–7 yrs. KFF Health News, May 2022 (same source as #28) (read 2026-09-29.) Second source: GAP.
  7. Overtime-gap illustration: 80-hr week value vs stipend — $170,414 vs $68,166; gap $102,248/yr. Site calculation from #1 + #7 + #17. (read 2026-09-29.)
  8. Bullock: UCSF resident; NEJM early-2020 account; ~5 days inpatient; quote. Medscape, Nov 1 2021 (read 2026-09-29.) Second source: GAP (NEJM full text not opened; Medscape quotes it).
  9. Amin: "Patients deserve doctors who aren't exhausted…" (UVM fellow). KFF Health News, May 2022 (same source as #28) (read 2026-09-29.) Second source: GAP.
  10. González: 80-hr ICU weeks; "preying on our altruism" (Harbor-UCLA chief/CIR president). KFF Health News, May 2022 (same source as #28) (read 2026-09-29.) Second source: GAP.
  11. CA 2026 min wage $16.90; NYC $17.00 (NOT used in copy — unverified). Resident Insider asserts; state pages not opened. Unverified — kept out of site copy. Second source: GAP — excluded from deck.

Parenting: leave & childcare rows 39–46

  1. ACGME leave floor: min 6 wks paid medical/parental/caregiver leave + 100% salary for first 6 wks (eff. 7/1/2022) — 6 wks / 100%. ACGME Answers: Resident Leave Policies (read 2026-09-30.) Second source: GAP.
  2. Surgery residents who took parental leave and took <2 wks; felt unsupported by peers / faculty (n=2,188, JAMA Surg 2019) — 42.5% / 30.4% / 32.7%. JAMA Surg parental-leave perceptions full text (PMC6686777) (read 2026-09-30.) Second source: GAP.
  3. Surgery residents correctly identifying ABS leave policy; top obstacles incl. no universal policy, strain on program — 3.8%. JAMA Surg parental-leave perceptions full text (PMC6686777) (same source as #40) (read 2026-09-30.) Second source: GAP.
  4. Ophthalmology residents taking parental leave: 44 of 283; median leave 4.5 wks (men 2 / women 6); no performance gap (JAMA Ophthalmol 2022) — 44/283; 4.5 wks. AMA summary of Huh et al. (read 2026-09-30.) Second source: GAP (JAMA page returned 403).
  5. Srikumaran (study PI): childbearing leave <6 wks "probably not enough"; supports up to 12 wks or hybrid return. AMA summary of Huh et al. (same source as #42) (read 2026-09-30.) Second source: GAP.
  6. Surgical programs leaving a single-parent resident with negative net income after expenses + childcare (295 programs, JAMA Network Open 2025) — 290/295 (98.3%). Mercante et al. full text (PMC11907318) (read 2026-09-30.) Second source: AMA, "Why child care costs have resident physician parents seeing red".
  7. Mean negative net income, parents of infants in West; mean infant care cost South–Northeast range — −$21,278; $9,694–$14,792. Mercante et al. full text (PMC11907318) (same source as #44) (read 2026-09-30.) Second source: Healio (regional means; West −$18,852).
  8. Surgical residents with / expecting children; mean surgical resident work week cited in same study — 28.6% / 84.3 hr/wk. Mercante et al. full text (PMC11907318) (same source as #44) (read 2026-09-30.) Second source: GAP.

Duty-hour honesty rows 47–49

  1. Residents reporting any noncompliance with 2011 duty-hour rules; admitting falsely reporting hours (n=6,202, JGME 2013) — 52.9% / 42.9%. Drolet et al. full text (PMC3771164) (read 2026-09-30.) Second source: GAP.
  2. Noncompliance by specialty (surgery highest) and as reported by program directors (n=549, 75.2% response) — 67.6% / 57.3%. Drolet et al. full text (PMC3771164) (same source as #47) (read 2026-09-30.) Second source: GAP.
  3. Interns reporting a duty-hour violation in ≥1 month of a yearlong study (Landrigan 2006, N=1,068, via Drolet) — 83.6%. Drolet et al. full text (PMC3771164) (same source as #47) (read 2026-09-30.) Second source: GAP.

Moonlighting rows 50–52

  1. Residents participating in moonlighting: non-orthopedic vs orthopedic (Johnson via 2023 systematic review, 21 studies, n=15,585) — 25.2% / 10.3%. Debt systematic review full text (PMC10502380) (read 2026-09-30.) Second source: GAP.
  2. Residents moonlighting to pay off debt (Carney ~half; Young 31%); larger debt → more likely to moonlight (Steiner, n=537) — 51–56% / 31%. Debt systematic review full text (PMC10502380) (same source as #50) (read 2026-09-30.) Second source: GAP.
  3. Moonlighting counts toward 80-hr limit; PGY-1 residents may not moonlight (ACGME CPR 6.25, via UW GME policy). UW GME Work Hours Policy (read 2026-09-30.) Second source: GAP.

Resident unions rows 53–56

  1. CIR membership today; largest housestaff union; SEIU local (vs ~31,000 in 2023 per Barger) — 40,000+. CIR official About page (read 2026-09-30.) Second source: Barger et al. full text (PMC11969282).
  2. Nonunion residents who would vote to unionize vs against (n=1,235; 20% already unionized; JAMA Network Open 2025) — 63% / <10%. Barger et al. full text (PMC11969282) (same source as #53) (read 2026-09-30.) Second source: GAP.
  3. Top factors in union vote (pay / hours); union vs nonunion salary with equal hours (57.3 vs 57.8 hr/wk) — 88% / 76%; $70,271 vs $65,455. Barger et al. full text (PMC11969282) (same source as #53) (read 2026-09-30.) Second source: GAP.
  4. Stanford 81% yes → 21% raise over 3 yrs; Mass General Brigham 75% yes, largest group (via Barger) — 81% / 75%. Barger et al. full text (PMC11969282) (same source as #53) (read 2026-09-30.) Second source: GAP.

Attrition & cost context rows 57–59

  1. Pooled attrition, general surgery residency (20+ studies, n=19,821); women vs men; leaving after PGY-1 (JAMA Surg review) — 18% / 25% vs 15% / 48%. Medical Design summary of Al-Omran JAMA Surgery review (read 2026-09-30.) Second source: GAP.
  2. Most common attrition cause (uncontrollable lifestyle); leavers going to another surgery program / anesthesia — top cause; 20% / 13%. Medical Design summary of Al-Omran JAMA Surgery review (same source as #57) (read 2026-09-30.) Second source: GAP.
  3. Burnout-attributable cost, CONTEXT ONLY: US physicians (Han, Ann Intern Med 2019) — not residents — $4.6B/yr; $7,600/physician. AMA summary (read 2026-09-30.) Second source: GAP — no resident-specific replacement-cost source found; do not present as resident figure.

Patient-safety evidence (early secondaries) rows 60–65

  1. Landrigan 2004 ICU RCT: more serious medical errors on 24-hr vs 16-hr schedule (20 interns, crossover, Brigham) — 36% more. KFF Health News (citing PubMed 15509817) (read 2026-09-30.) Second source: Critical Care evidence review full text (PMC1175930).
  2. Lockley 2004 (same trial): extra sleep per week and attentional lapses on intervention schedule; LOS/mortality unchanged — +5.8 hr/wk; half the lapses. Critical Care evidence review full text (PMC1175930) (same source as #60) (read 2026-09-30.) Second source: GAP.
  3. Block 2014: no significant differences after 2011 reform (LOS, 30-day readmit, mortality, HACs, ICU admits) — no sig. diff.. AHRQ PSNet summary (read 2026-09-30.) Second source: GAP.
  4. FIRST trial: surgical programs randomized; flexible hours noninferior for death/serious complications; residents preferring standard rules — 117 programs; 14%. Feinberg News Center (read 2026-09-30.) Second source: Queen's Univ. duty-hours review.
  5. iCOMPARE: IM programs randomized; 30-day mortality noninferior under flexible rules; flexible interns more dissatisfied; burnout >66% both arms — 63 programs. Queen's Univ. duty-hours review (same source as #63) (read 2026-09-30.) Second source: GAP (NEJM full text paywalled; conclusion corroborated by NEJM abstract page snippet only — not cited).
  6. Sen survey: self-reported serious errors before vs after 2011 rule (n=2,323 interns, 51 residencies); Desai RCT: more sleep, fewer conferences — 20% → 23%. KFF Health News (citing PubMed 15509817) (same source as #60) (read 2026-09-30.) Second source: GAP.

Pillar 1: patient-safety primaries rows 66–78

  1. Landrigan 2004 UPGRADE (direct PubMed): 35.9% more serious medical errors, traditional vs intervention schedule (136.0 vs 100.1 per 1,000 patient-days, P<0.001); 2,203 patient-days, 634 admissions — 35.9%. PubMed PMID 15509817 (Landrigan et al., N Engl J Med 2004;351:1838-48) (read 2026-09-30 via NCBI E-utilities.) Second source: Replaces KFF-secondary row 60 as primary.
  2. Same trial, error subtypes: 56.6% more nonintercepted serious errors; unit-wide serious-error rate 22.0% higher (193.2 vs 158.4); 20.8% more serious medication errors (P=0.03); 5.6x serious diagnostic errors (18.6 vs 3.3, P<0.001) — 56.6% / 22.0% / 20.8% / 5.6x. PubMed PMID 15509817 (Landrigan et al., N Engl J Med 2004;351:1838-48) (same source as #66) (read 2026-09-30.) Second source: GAP.
  3. Lockley 2004 UPGRADE (direct PubMed): traditional 84.9 vs intervention 65.4 hr/wk (19.5 fewer, P<0.001); +5.8 hr sleep/wk (P<0.001); less than half the attentional-failure rate on on-call nights (P=0.02); n=20 interns — +5.8 hr; <1/2 lapses. PubMed PMID 15509816 (Lockley et al., N Engl J Med 2004;351:1829-37) (read 2026-09-30 via NCBI E-utilities.) Second source: Replaces secondary row 61 as primary.
  4. Barger 2005 drowsy driving: crash OR 2.3 (1.6–3.3) and near-miss OR 5.9 (5.4–6.3) after extended (≥24 hr) shift; each added extended shift/month +9.1% crash risk, +16.2% commute-crash risk; n=2,737 interns, 17,003 reports — OR 2.3 / 5.9. PubMed PMID 15647575 (Barger et al., N Engl J Med 2005;352:125-34) (read 2026-09-30 via NCBI E-utilities.) Second source: GAP.
  5. Ayas 2006 needlesticks: percutaneous-injury OR 1.61 extended vs nonextended work (1.31 vs 0.76 per 1,000 opportunities); top contributing factors lapse in concentration 64%, fatigue 31% (of 448 injuries); 29.1 vs 6.1 consecutive hrs — OR 1.61. PubMed PMID 16954484 (Ayas et al., JAMA 2006;296:1055-62) (read 2026-09-30 via NCBI E-utilities.) Second source: GAP.
  6. Philibert 2005 meta-analysis: 60 studies (959 physicians, 1,028 nonphysicians, 5,295 effect indexes); <30 hrs sleep loss cut physicians' overall performance ~1 SD and clinical performance >1.5 SD — ~1 SD / >1.5 SD. PubMed PMID 16335329 (Philibert, Sleep 2005;28:1392-402) (read 2026-09-30 via NCBI E-utilities.) Second source: GAP (author at ACGME; conclusion notes current limits "may not completely guard" performance).
  7. Fahrenkopf 2008: depressed residents 6.2x medication errors per resident-month (1.55 vs 0.25, P<0.001); burnt-out vs non-burnt-out similar (0.45 vs 0.53, P=0.2); 20% depressed, 74% burnt out; n=123 peds residents, 45 errors by active surveillance — 6.2x; burnout null. BMJ 2008;336:488-91 full text (PMC2258399) (read 2026-09-30.) Second source: PubMed PMID 18258931.
  8. West 2009: among IM residents, positive depression screen → OR 2.56 subsequently reported major medical error (1.76–3.72); burnout subscales, fatigue, sleepiness also associated; 39% reported ≥1 major error; n=380 — OR 2.56. PubMed PMID 19773564 (West et al., JAMA 2009;302:1294-300) (read 2026-09-30 via NCBI E-utilities.) Second source: GAP (errors self-perceived, longitudinal design).
  9. Tawfik 2018: physicians with burnout OR 2.22 self-reported major error in prior 3 months (1.79–2.76), adjusted for specialty/hours/fatigue/safety grade; fatigue OR 1.38; n=6,586, 10.5% reported error — OR 2.22. PubMed PMID 30001832 (Tawfik et al., Mayo Clin Proc 2018;93:1571-80) (read 2026-09-30 via NCBI E-utilities.) Second source: GAP (attending physicians, self-reported errors — scope, not resident-measured).
  10. FIRST UPGRADE (direct PubMed): 117 surgery programs, 138,691 patients; death/serious complications 9.1% flexible vs 9.0% standard (noninferiority met); flexible residents less often left operations (7.0% vs 13.2%) or handed off (32.0% vs 46.3%) — 9.1% vs 9.0%. PubMed PMID 26836220 (Bilimoria et al., N Engl J Med 2016;374:713-27) (read 2026-09-30 via NCBI E-utilities.) Second source: Replaces Feinberg-secondary row 63 as primary.
  11. iCOMPARE-patient UPGRADE (direct PubMed): 63 IM programs; change in 30-day mortality noninferior (flexible 12.5% trial vs 12.6% pretrial; standard 12.2% vs 12.7%; margin 1 pp, upper 1-sided 95% CI 0.93%, P=0.03) — noninferior, CI 0.93%. PubMed PMID 30855740 (Silber et al., N Engl J Med 2019;380:905-14) (read 2026-09-30 via NCBI E-utilities.) Second source: Replaces review-secondary row 64 (mortality half) as primary.
  12. iCOMPARE-education (direct PubMed): flexible-program interns more dissatisfied — well-being OR 2.47 (1.67–3.65), educational quality OR 1.67; in-training exam 68.9% vs 69.4%, noninferiority margin 2 pp NOT met (P=0.06) — OR 2.47; exam P=0.06. PubMed PMID 29557719 (Desai et al., N Engl J Med 2018;378:1494-508) (read 2026-09-30 via NCBI E-utilities.) Second source: PMCID PMC6101652 (free full text, abstract read here).
  13. Block 2014 UPGRADE (direct PubMed): single academic center, diff-in-diff resident vs hospitalist services, pre (2008–2011) vs post (2011–2012): no significant differences in LOS, 30-day readmit, ICU admit, inpatient mortality, HACs — no sig. diff.. PubMed PMID 24677678 (Block et al., J Hosp Med 2014;9:347-52) (read 2026-09-30 via NCBI E-utilities.) Second source: Replaces PSNet-secondary row 62 as primary; single-center + first-year-only caveat.

Pillar 2: the money trail rows 79–104

  1. Medicare GME mechanism (history): 1965 Congress approved GME in teaching hospitals' "reasonable costs" — "including stipends of trainees" — letting Medicare payments "partially recompense salaries and benefits for house officers, administrative costs for GME, and cost of faculty". "Medicare Financing of Graduate Medical Education" full text (PMC1495035) (read 2026-09-30.) Second source: GAP (describes cost-based era; current PPS-era mechanism = rows 80–83).
  2. Medicare GME FY2023 (est.): $21.2B total = $6.1B DGME + $15.0B IME; DGME FTEs 112,230 (allopathy/osteopathy) + 4,201 (podiatry/dentistry); IME FTEs 119,328 + 4,706 — $21.2B; $6.1B / $15.0B. CRS R48636, "Federal Support for Graduate Medical Education" (Aug 19, 2025) (read 2026-09-30.) Second source: GAP.
  3. Medicare per-resident amount (PRA) FY2023 (est.): average $133,000 primary care / $131,000 non-primary; maximum $307,000 / $290,000; minimum $17,000 / $14,000 — $133k / $131k avg. CRS R48636, "Federal Support for Graduate Medical Education" (Aug 19, 2025) (same source as #80) (read 2026-09-30.) Second source: GAP (PRA ≠ stipend: covers salaries + supervision + overhead, Medicare share only).
  4. DGME formula (CMS): hospital-specific per-resident amount × FTE residents × Medicare patient share; PRA rooted in base-year direct GME costs "including teaching physician and resident salaries"; IME = IPPS percentage add-on per Medicare discharge for higher indirect costs. CMS, "Medicare Policy Clarifications on GME Payments" (4/8/2005) (read 2026-09-30.) Second source: 42 CFR §413.77 PRA methodology (cited, not opened — regulation text, use CMS summary as claim source).
  5. Federal GME context: all-program federal GME ≈$29B (most recent year available); last comprehensive estimate FY2012 ≈$16B; Medicare alone FY2023 $21.2B (est.) — ≈$29B. CRS R48636, "Federal Support for Graduate Medical Education" (Aug 19, 2025) (same source as #80) (read 2026-09-30.) Second source: GAP.
  6. Resident stipends taxable: IRS — "You must include in gross income everything you receive in payment for personal services. In addition to wages, salaries, commissions, fees, and tips…" (narrow claim: stipends for services = taxable compensation). IRS Publication 525 (2025), Taxable and Nontaxable Income (read 2026-09-30.) Second source: GAP (no claim about specific exclusions/deductions beyond this sentence).
  7. Resident FICA: IRS longstanding position — "medical residents are not students… they are full-time employees" ineligible for student FICA exception; "services provided as a medical resident on or after April 1, 2005 are subject to FICA tax". IRS private-determination letter 09-0071 (read 2026-09-30 via PDF text extraction.) Second source: GAP (FICA history limited to this letter; Mayo Foundation SCOTUS decision deliberately not cited — not IRS text).
  8. Student-FICA test: exception only where education predominates over employment in the relationship; cites TD 9167 final regs (12/21/2004) + Rev Proc 2005-11 safe harbor. IRS, "Student exception to FICA tax" (read 2026-09-30.) Second source: IRS private-determination letter 09-0071.
  9. AAMC first-year mean stipend 2020→2023: $58,921 → $63,800 (+8.3% nominal); CPI-U 257.8 → 305.1 (+18.4%); CPI-adjusted (1969 $) $7,931 → $7,256 (−8.5% real) — −8.5% real. AAMC 2023 Survey of Resident/Fellow Stipends and Benefits Report, Table 2 (official AAMC report PDF, institutional mirror) (read 2026-09-30.) Second source: GAP (same-table comparison; AAMC warns respondent cohorts differ year to year).
  10. AAMC long view: 2023 CPI-adjusted first-year mean $7,256 = "17.0% real increase over the $6,200 mean… in 1968-1969"; 2022→2023 CPI-adjusted $7,137 → $7,256 (+$119) — +17.0% since 1968. AAMC 2023 Survey of Resident/Fellow Stipends and Benefits Report, Table 2 (official AAMC report PDF, institutional mirror) (same source as #87) (read 2026-09-30.) Second source: GAP.
  11. AAMC historical anchors (first-year mean, CPI-U): 2019-20 $57,191 (256.1); 2014-15 $51,586 (238.3); 2004-05 $40,788 (189.7); 2018-19 $56,126 (252.0); 2016-17 $54,127 (241.0). AAMC 2019-2020 RFSB Report, Table 2 (1968-69→2019-20) (read 2026-09-30.) Second source: AAMC 2023 Survey of Resident/Fellow Stipends and Benefits Report, Table 2 (official AAMC report PDF, institutional mirror).
  12. BLS CPI-U (CUUR0000SA0, via public API): 2024 annual avg 313.7 (mean of 12 monthly values 308.417–315.605); 2025 partial avg 321.9 (11 months; October missing/dash in BLS file) — 313.7 / 321.9p. BLS Public API v2, series CUUR0000SA0, retrieved 2026-09-30 (read 2026-09-30.) Second source: GAP (2025 = partial-year; October 2025 value absent from BLS response).
  13. Real-stipend arithmetic 2019-20→2025: $57,191 → $68,166 = +19.2% nominal; CPI-U 256.1 → 321.9p = +25.7%; real change 1.192/1.257 − 1 = −5.2% — −5.2% real. Site calculation from #89 + #1 + #90. (read 2026-09-30.) Second source: AAMC-internal −8.5% (row 87) is the cleaner same-table figure; this extends to 2025 with stated partial-year CPI.
  14. Marcus/CORR: mean CEO comp at 22 major nonprofit centers $1.6M → $3.1M (+93%), CFO $740K → $1.4M (+83%), 2005→2015, inflation-adjusted (2015 CPI); IRS-990-based — +93% / +83%. Du/Rascoe/Marcus, Clin Orthop Relat Res 2018;476:1910-19, full text (PMC6259823) (read 2026-09-30.) Second source: PubMed PMID 30001293.
  15. Marcus wage gaps 2005→2015: CEO vs orthopaedic surgeons 3:1 → 5:1; vs pediatricians 7:1 → 12:1; vs RNs 23:1 → 44:1; CFO vs orthopaedic 1.5:1 → 2.2:1 — 3:1→5:1 etc.. Du/Rascoe/Marcus, Clin Orthop Relat Res 2018;476:1910-19, full text (PMC6259823) (same source as #92) (read 2026-09-30.) Second source: GAP (site-specific exec pay vs national-mean clinician pay — paper's stated method).
  16. Marcus national (BLS-based) 2005→2015: healthcare mean wages +8% ($50,435→$54,618); management +14% ($93,202→$105,801); physicians +10% ($187,895→$206,242); workforce 13M→15M; 2015 staffing per physician: 10 nonclinical + 1 management + 14 nonphysician clinical; wage-burden $663B→$865B (+30%) — +8% / +14% / +10%. Du/Rascoe/Marcus, Clin Orthop Relat Res 2018;476:1910-19, full text (PMC6259823) (same source as #92) (read 2026-09-30.) Second source: GAP.
  17. Jenkins/Med Care: nonprofit-hospital CEO mean comp +34% ($1.0M → $1.3M in 2019 dollars) vs RN mean wages +2.3% ($75,652 → $77,460) 2012→2019; n=1,047 systems/hospitals (2012), 812 (2019); IRS-990-based — +34% vs +2.3%. Jenkins/Short/Ho, Med Care 2025;63:787-93, full text (PMC12422613) (read 2026-09-30.) Second source: PubMed PMID 40846652.
  18. Jenkins: quality-pay link weaker in 2019 than 2012 — "CEOs are being rewarded more for leading large hospitals or systems, but not for providing higher quality care". Jenkins/Short/Ho, Med Care 2025;63:787-93, full text (PMC12422613) (same source as #95) (read 2026-09-30.) Second source: GAP.
  19. BLS OEWS May 2025 (via public API, latest): NP mean $137,300 / median $132,300 (323,040 employed); PA mean $141,280 / median $135,880 (162,150); RN mean $101,420 / median $97,550 (3,379,720). BLS Public API v2, series OEUN000000000000029117104/113, …29107104/113, …29114104/113, retrieved 2026-09-30 (read 2026-09-30.) Second source: GAP (BLS web pages bot-blocked 403; API is the readable official source; SOC 29-1171/29-1071/29-1141).
  20. Same-year annual-pay ratios, arithmetic: NP mean / PGY-1 mean $137,300/$68,166 = 2.01x; PA $141,280/$68,166 = 2.07x; RN $101,420/$68,166 = 1.49x (all 2025) — 2.01x / 2.07x / 1.49x. Site calculation from #97 + #1. (read 2026-09-30.) Second source: Weighting caveat: OES means are employment-weighted national; AAMC mean is unweighted institutional PGY-1 — levels comparable, ratio approximate.
  21. PA wage growth 2000→2013: +40% vs cumulative inflation 35.3%; 2013 clinical-role median $44.70/hr (BLS datasets, peer-reviewed); linear projection to 2025 (projection NOT cited as fact) — +40% vs 35.3%. PubMed PMID 25989436 (Quella/Brock/Hooker, JAAPA 2015;28:56-63) (read 2026-09-30 via NCBI E-utilities.) Second source: GAP (abstract-level read; 2025 projection excluded as forecast, not evidence).
  22. Health-workforce real wage growth 2001→2017 (ACS-based, peer-reviewed): nurses (incl. NPs per methods) cumulative median +9.92%; healthcare practitioners +5.68%; physicians +37.6%; support <1%; direct-care −8.4% — +9.92% / +5.68% / +37.6%. "Real wage growth in the U.S. health workforce," Hum Resour Health 2021, full text (PMC8403397) (read 2026-09-30.) Second source: PubMed PMID 34454538.
  23. 2003 ACGME rules (quoted as "current requirements" in ACGME doc): continuous duty ≤24 hrs + up to 6 additional (=30); no new patients after 24 hrs; in-house call ≤ every 3rd night; 1 day in 7 free incl. call; internal moonlighting counts toward 80 (88 w/ exception) — 24+6=30. ACGME, "Final Report on Cost Analysis for 2011 CPRs" (VI.G comparison table + survey quotes of 2003 text) (read 2026-09-30.) Second source: GAP (2003 text via ACGME's own quotation, not the 2003 CPR PDF itself).
  24. 2011 ACGME rules (VI.G): PGY-1 duty ≤16 hrs; PGY-2+ ≤24 continuous + ≤4 additional (=28, down from 30); all moonlighting counts toward 80; PGY-1 no moonlighting; ≥1 day free/week avg 4 wks, no at-home call on free days; strategic napping "strongly suggested" — 16 / 28. ACGME, "Final Report on Cost Analysis for 2011 CPRs" (VI.G comparison table + survey quotes of 2003 text) (same source as #101) (read 2026-09-30.) Second source: ACGME JGME 2011 rationale supplement (context only).
  25. 2017 ACGME revision: PGY-1 cap returns to 24 (+4 transitions) — "same schedule as other residents"; 80-hr week, 1-in-7, q3 call UNCHANGED; effective 7/1/2017 — 24+4 for all. ACGME press release 3/10/2017 (read 2026-09-30.) Second source: ACGME newsroom release (same content).
  26. What barely changed 2003→2026: 80-hr week, every-3rd-night call limit, and 1-day-in-7 free are constant across all four eras; only the PGY-1/extended-shift cap moved (30 → 16 → 28) plus the +6→+4 handoff narrowing. Site calculation from #101 + #102 + #103 + #8 + #9 + #11. (read 2026-09-30.) Second source: 2026 CPR keeps 24+4 for all incl. PGY-1 (row 9) — the 2017 structure persists.

Found an error?

Tell us the page, the line, and the better source. We fix cited errors promptly.

Read the Recommendations