In a typical first-round slate of eight, two or three healthcare cases die on the same mistake: the candidate optimizes beds, or "volume," before naming how the client gets paid. Under Medicare's inpatient prospective payment system, payment is usually a single rate per stay by diagnosis-related group, not per day. Extra days are mostly cost. Commercial contracts often pay more than twice Medicare for the same services. If you treat the hospital like a hotel filling rooms, you will recommend the wrong wing.
This guide stays on the provider and payer side of healthcare interviews—hospitals, health systems, physician groups, and health plans. Drug launch and patent-cliff math live on the pharma case interview page; mixed pipelines belong on life sciences. After this piece you should be able to pick the payment unit in the first minute, split contribution by payer, and finish a throughput decision without inventing a new building.
Start with the payment unit, not the oak tree
Ask, before you draw anything: is revenue a DRG (or similar bundle) per discharge, a per diem, a fee schedule, or a capitated / premium dollar with a medical loss ratio underneath?
CMS pays most acute inpatient Medicare cases under IPPS: each stay maps to an MS-DRG with a relative weight, then a standardized rate adjusted for wages, teaching, DSH, outliers, and quality programs. The FY 2025 IPPS final rule continues that architecture—one payment for the stay, not a meter that runs while the patient occupies a bed (CMS fact sheet). That is why length of stay is a margin lever under Medicare even when occupancy looks fine.
Commercial is different. KFF, citing RAND's employer claims work, reports private insurance paid about 267% of Medicare on average for hospital inpatient and outpatient facility services in 2022. RAND's own Round 5.1 summary puts the all-services average near 254% of Medicare for 2022 (RAND). In the room, you do not need the exact multiple. You need to refuse a blended "average payment per discharge" until the exhibit is split by payer.
If the client is a payer, flip the P&L. Premium is the top line; medical cost is the big variable. Under the ACA, fully insured large-group plans must hit an 85% medical loss ratio floor (80% in individual and small group), or rebate the shortfall (CMS). UnitedHealth Group's 2024 Form 10-K reported a medical care ratio of 85.5%, up from 83.2% in 2023—useful as a public illustration that plan economics live in a few points of MCR, not in "synergies" (SEC filing).
The metrics interviewers expect you to touch
You will not get every line in every case. Touch the ones that change the recommendation.
| Metric | Why it shows up | What "good" looks like in an answer |
|---|---|---|
| Payer mix | Same DRG, different contribution | Split volume and margin by commercial / Medicare / Medicaid / other |
| Reimbursement vs cost | DRG rate is not chargemaster | Net payment, not list charges; say so out loud |
| Gross-to-net | Hospitals bill high, collect less | Contractual adjustments and denials sit between gross charges and net patient revenue |
| Occupancy | Capacity signal, not demand | Staffed beds, not licensed; by unit if the exhibit allows |
| ALOS / discharges | Throughput under fixed beds | Discharges = patient-days / ALOS; occupancy can rise while discharges fall |
| MLR / MCR (payer cases) | Plan margin | Medical cost / premium; admin and underwriting are the residual |
National context keeps your numbers honest. In 2023, private insurance accounted for 37% of U.S. hospital spending, Medicare 25%, and Medicaid 19% (KFF). HCA Healthcare's 2024 Form 10-K shows how a large system actually reports volume: 73% occupancy, average length of stay 4.8 days, and managed care plus other insurers at roughly half of revenues (HCA 10-K). Those are not case inputs. They are the range an interviewer finds plausible.
UK prompts swap DRGs for tariffs and block contracts, but the occupancy logic is the same. NHS England publishes overnight bed availability and occupancy quarterly via KH03; recent Nuffield Trust summaries of that series put general and acute overnight occupancy near 92–93%—a system that is bed-constrained in a way a 70% U.S. community hospital often is not (NHS England bed data).
Occupancy lies until you convert it to discharges
Patient-days = staffed beds × occupancy × 365. Discharges = patient-days / ALOS. Adding beds grows patient-days only if you can staff and fill them. Cutting ALOS grows discharges with the beds you already have.
Candidates fail this conversion in two directions. They see 72–78% occupancy at a U.S. system and announce a capacity crisis—HCA ran about 73% in 2024 and still grew admissions. Or they see NHS-style 90%+ occupancy and talk about "marketing for volume" as if empty rooms were the constraint. Name the bottleneck: ICU vs med/surg, weekday vs weekend staffing, or discharge destination (SNF, home health), not a headline occupancy percentage.
Gross-to-net is the other silent trap. Chargemaster prices are not revenue. Net patient revenue is what remains after contractual allowances, charity care, and bad debt. If an exhibit shows "charges," ask for net by payer before you multiply anything.
Worked mini-case: Cedar Ridge's wing, its ALOS program, and 400 commercial shifts
Prompt (invented figures, internally consistent). Cedar Ridge Medical Center has 220 staffed beds, 78% occupancy, and ALOS of 4.9 days. Variable cost averages $10,600 per discharge. Net payment and mix:
| Payer | Share of discharges | Net payment / discharge | Contribution / discharge |
|---|---|---|---|
| Commercial | 30% | $22,000 | $11,400 |
| Medicare | 45% | $11,200 | $600 |
| Medicaid | 20% | $8,800 | -$1,800 |
| Self-pay / other | 5% | $2,800 | -$7,800 |
The CFO offers three options for next year's operating margin: (A) a $6.5m, 24-bed med/surg wing needing $0.9m extra annual nursing, assumed to fill at system occupancy and mix; (B) a discharge-planning program that cuts ALOS to 4.6 days for $0.35m and no capital; (C) an ambulatory-surgery and referring-physician push that converts 400 Medicare discharges to commercial at the same variable cost. Quality leadership will kill any option that raises 30-day readmissions. Recommend one primary move.
Structure. Current discharge factory → contribution by payer → ALOS lever holding beds and occupancy → bed-add lever → mix shift → capital and quality constraint.
Math. Patient-days today = 220 × 0.78 × 365 = 62,634. Discharges = 62,634 / 4.9 ≈ 12,782.
Blended contribution per discharge = 0.30×11,400 + 0.45×600 + 0.20×(-1,800) + 0.05×(-7,800) = 3,420 + 270 − 360 − 390 = $2,940.
Option B (ALOS 4.6, same beds and occupancy): discharges = 62,634 / 4.6 = 13,616, or +834 stays. Extra contribution ≈ 834 × $2,940 ≈ $2.45m, minus $0.35m program cost → about $2.1m operating-profit lift, zero capital.
Option A (24 beds at 78% and 4.9 ALOS): extra patient-days = 24 × 0.78 × 365 = 6,833; extra discharges ≈ 1,394; extra contribution ≈ $4.1m, minus $0.9m nursing → about $3.2m year-1 P&L—but only after writing a $6.5m check, only if those beds fill, and only if mix does not worsen. At 78% the hospital is not short of rooms in the U.S. sense; it may be short of turns.
Option C (400 Medicare → commercial): incremental contribution per stay = 11,400 − 600 = $10,800. Times 400 = $4.32m, with no new beds and no ALOS heroics. Feasibility is the issue—contracts, ASC capacity, and physician alignment—not the arithmetic.
Recommendation. Lead with C if the interviewer confirms referral and OR capacity can absorb 400 shifts without price concessions that erase the gap. Run B in parallel as the no-regret throughput program and watch readmissions weekly. Park A until med/surg occupancy actually binds after ALOS and mix improve. Next exhibit you want: contribution and ALOS by DRG and by payer, not a blended average.
Ned's rule. In a healthcare provider case, name the payment unit in the first thirty seconds. If it is a DRG-style bundle, length of stay is a margin lever; if you cannot split payer mix, you do not yet have a volume story.
What the scoring sheet rewards (and what it ignores)
The sheet rewards a clear objective (margin, wait time, build vs buy capacity), the beds-to-discharges conversion, a payer-mix so-what, and an explicit quality constraint. It does not reward naming Porter's Five Forces, inventing a "healthcare framework," or claiming you know a firm's confidential industry playbook. Firms publish practice-area pages; they do not publish the case you are about to see. Treat Reddit and Glassdoor as folklore warehouses, not as primary sources.
A generic profitability tree misses the binding mechanics: volume is discharges, price is contract × weight × base rate, variable cost has a per-day component under bundled payment, and fixed cost is staffing a unit. EMTALA still constrains the fantasy of "closing the unprofitable ED"; you can change admission criteria and site of care, not abandon the door.
Practice the conversion under time pressure
Run one provider-style case today and force yourself to say the payment unit before you sketch volume. Use /start for a five-minute typed first rep with instant scores, or /interview for the live voice case with Ned (~18 minutes, seven-score debrief). For the discharge math alone, /drills/math and /drills/quick-math (free forever) are enough. Everything is open for seven days, no card; then $120 for a recruiting season or $49 a month.
CoachNed is independent and unaffiliated with McKinsey, BCG, Bain, Deloitte, or any other firm named for interview practice.
Frequently asked questions
What is a healthcare case interview usually about?
Most firm cases labeled "healthcare" are provider economics (capacity, service-line mix, site of care, payer mix) or payer economics (premium, medical cost, MLR). Pharma launch and pipeline cases are a different product; clarify the client type in your opening questions.
How do Medicare DRGs change the math?
Under IPPS, Medicare typically pays a prospective amount for the stay based on MS-DRG weight and hospital adjustments. Holding the DRG fixed, a longer stay usually raises cost without raising payment, so ALOS and discharge planning sit in the recommendation, not as afterthoughts.
What occupancy rate should I assume is "full"?
There is no universal threshold. Large U.S. systems often run in the low-to-mid 70s on an annual average; NHS acute overnight occupancy often sits above 90%. Ask whether the exhibit is staffed beds, whether ICU is the bottleneck, and whether demand is seasonal before you call for more concrete.
How is a payer case different from a hospital case?
A hospital case is mostly net revenue per discharge or visit minus cost to deliver care. A payer case starts from premium PMPM, medical cost PMPM, and the MLR or medical care ratio, with admin and risk as the residual. Confusing the two is an immediate tell.
Should I use a standard profitability framework?
Use the logic—revenue, cost, volume, price—but rebuild the tree around payment unit, payer mix, and beds vs discharges. Naming a canned framework rarely helps; showing the conversion and the mix split does.
Sources
- CMS Acute Inpatient PPS overview — MS-DRG payment architecture under IPPS. Checked 2026-09-24.
- CMS FY 2025 IPPS / LTCH PPS final rule fact sheet — current-year inpatient payment system context. Checked 2026-09-24.
- KFF Key Facts About Hospitals — 2023 hospital spending by payer and commercial prices vs Medicare. Checked 2026-09-24.
- RAND Hospital Price Transparency Study Round 5 — commercial hospital prices ~254% of Medicare in 2022. Checked 2026-09-24.
- CMS Medical Loss Ratio fact sheet — 80%/85% MLR floors for fully insured markets. Checked 2026-09-24.
- UnitedHealth Group 2024 Form 10-K (SEC) — reported medical care ratio of 85.5% for 2024. Checked 2026-09-24.
- HCA Healthcare 2024 Form 10-K (SEC) — occupancy, ALOS, and payer revenue mix for a large U.S. hospital system. Checked 2026-09-24.
- NHS England bed availability and occupancy (overnight) — primary KH03 occupancy series for England. Checked 2026-09-24.