Post

Conversation

Alex Armlovich
@aarmlovi
One of the trickiest open questions in healthcare econ: -What is the average hospital's Marginal Revenue Product from a new resident? (paid <$100k, less than a traveling nurse) -If the MRP of a resident exceeds their wage, why don't hospitals self-fund more residency slots?
Quote
Lawson Mansell
@lawsonhmansell
The doctor shortage is not due to lack of interest in becoming a doctor. It's due to a lack of slots. x.com/DrIanWeissman/…
Image
12.7K
Views
David Watson 🥑
Post your reply

Alex Armlovich
@aarmlovi
Grok's pro forma exercise here is pretty good:
Are there any estimates of what the MRP of an MD resident at a typical hospital anywhere in the US?
Answer by Grok, expand to see full conversation
Estimating the Marginal Revenue Product (MRP) of a medical resident (MD resident) at a typical U.S. hospital is a challenging task due to the complexity of healthcare economics, limited public data, and the unique dual role residents play as both trainees and productive workers. However, we can piece together an informed estimate based on available research, industry data, and the context provided by the X posts and related web results as of 01:34 PM EDT on Wednesday, July 9, 2025. Let’s break this down step-by-step, leveraging my curiosity as Grok 3 and the latest insights. Key Components of MRP Recall that MRP is calculated as: \[ MRP = MPP \times MR \] MPP (Marginal Physical Product)**: The additional healthcare services (e.g., patient visits, procedures, or hours of care) a resident contributes. MR (Marginal Revenue)**: The revenue generated from those services, influenced by reimbursement rates (e.g., Medicare, private insurance) and hospital pricing. For residents, MPP is tricky because their work is supervised, and they split time between learning and patient care. MR depends on the hospital’s payer mix and local market rates. Available Evidence and Estimates Resident Wages as a Baseline According to the web result from Inspira Advantage (updated April 10, 2025), medical residents earn an average of $60,000 to $70,000 annually in their first year, with total compensation (including benefits) often nearing $100,000 when accounting for overtime and institutional support. This aligns with Alex Armlovich’s mention of residents being paid less than $100,000. In a competitive market, wages should approximate MRP if labor supply meets demand. However, the doctor shortage and capped residency slots suggest MRP might exceed wages due to market distortions. Productivity Contributions The European Journal of Health Economics article highlights that residents contribute to healthcare services while training. Studies suggest their productivity varies by specialty and hospital type. For example, a 2007 Arch Surg study (cited in the ACS web result) estimated that residents handle a significant portion of routine care, potentially equivalent to 0.5 to 0.8 full-time equivalent (FTE) attending physicians in teaching hospitals. A 2013 analysis (Medicare Payments for Graduate Medical Education) estimated that residents perform about 2,000 to 3,000 patient care hours annually, depending on specialty and program intensity. Revenue Generation Reimbursement rates provide a clue to MR. Medicare pays hospitals for resident services under Graduate Medical Education (GME) funding, with direct medical education (DME) payments averaging $100,000 to $150,000 per resident-year (per 2013 NCBI data, adjusted for inflation to ~$130,000–$200,000 in 2025 dollars). Indirect medical education (IME) adjustments add another $50,000 to $100,000 per resident due to teaching costs and patient complexity. Private insurance often reimburses at higher rates—e.g., $150 to $300 per outpatient visit or $500 to $1,000 per procedure, depending on specialty (e.g., surgery vs. family medicine). If a resident handles 10–15 visits or 5–10 procedures weekly, annual revenue could range from $100,000 to $300,000, net of supervision costs. Studies on Resident MRP A 2007 Arch Surg study (Mitchell et al., cited in the ACS web result) modeled staffing needs and suggested that replacing residents with midlevel practitioners (e.g., nurse practitioners) could cost hospitals $150,000 to $200,000 per FTE due to higher wages and lower productivity. This implies residents’ MRP might be in a similar range, given their lower cost. The European Journal of Health Economics notes that resident productivity can be positive or negative depending on supervision demands. When positive, MRP likely exceeds their wage, supporting Alex’s hypothesis that hospitals could benefit from more slots. Synthesizing an Estimate Let’s construct a rough estimate for a “typical” U.S. hospital (e.g., a medium-sized teaching hospital with a mix of Medicare and private payers): MPP: Assume a resident contributes 2,500 patient care hours annually (mid-range estimate), equivalent to 0.6–0.8 FTE of an attending physician’s productive time, adjusted for supervision. Attending physicians generate **$300,000 to $500,000 in annual revenue (per 2023 MGMA data on physician productivity), so a resident’s MPP might translate to $180,000 to $400,000 in potential service value. MR: With a payer mix of 40% Medicare ($130,000–$200,000 per resident via DME/IME) and 60% private insurance ($150,000–$300,000 from visits/procedures), the average MR per resident could range from **$140,000 to $250,000, net of overhead (e.g., 50% of gross revenue covers costs). MRP Range: Multiplying MPP contribution by MR, the MRP likely falls between **$150,000 and $350,000 per resident-year, with variability by specialty (e.g., higher for surgery, lower for primary care). Challenges and Variability Specialty Differences**: Surgical residents may generate higher MRP ($250,000–$400,000) due to procedure revenue, while family medicine residents might be closer to $100,000–$200,000. Hospital Type**: Urban teaching hospitals with complex cases may see higher MRP than rural community hospitals. Data Gaps*: No single study provides a definitive MRP figure. The ACS article (Reframing the Value of Resident Education) suggests the cost of *not training residents (e.g., hiring alternatives) exceeds $200,000 per slot, implying MRP is at least that high. Regulatory Constraints**: As noted in the X posts, ACGME caps and Medicare funding limits prevent hospitals from capturing full MRP, distorting market signals. Conclusion Based on the available data, the MRP of an MD resident at a typical U.S. hospital is likely in the range of $150,000 to $350,000 annually, with a midpoint estimate of around $250,000. This exceeds their wage (~$100,000), supporting Alex Armlovich’s question about why hospitals don’t fund more slots—suggesting barriers like funding caps or accreditation limits outweigh economic incentives. This estimate is speculative due to data scarcity, but it aligns with economic theory and the X thread’s focus on residency shortages. For a precise figure, hospital-specific studies or updated GME cost analyses (post-2023) would be needed. What do you think—should we dig deeper into specialty-specific data or explore hospital financial reports for more clues? I’m eager to explore further with you!
1.1K
Alex Armlovich
@aarmlovi
Healthcare is the funniest industry It gobbles 1 in 5 dollars in the richest country in human history, yet every single participant in the US healthcare industry is like "whoa, hey, don't look at me I'm losin' munny ova here"
Quote
Seth Borman
@SethBorman
Replying to @aarmlovi @arpitrage and @varma_ashwin97
Physician services are, themselves, a loss leader. A hospitalist needs to see at least 30-40 patients a day to pay their own salary. They actually see 18. A doctor with 3 residents being paid the same as him would need to see 60-80 pts a day.
1.3K
Seth Borman
@SethBorman
If you're deeply incurious about how things work I would suggest simply not asking on the internet. There's a reason that most physicians work for hospitals, the revenue from high priced hospital services covers the losses from physicians services.
206
Alex Armlovich
@aarmlovi
I'm not assigning specific blame I'm just noting an amusing dynamic in healthcare discourse! It is insane that US reimbursements & US costs are so high, and even super-dense, super-rich places like NYC are seeing hospital closures. It's genuinely remarkable
194
Alex Armlovich
@aarmlovi
Hospitals are always saying "oh we don't get enough residency slot funding from Congress" And I'm like yeah...nurses don't get training slots funded by Congress either! Unless residents are somehow massive money-losers at delivering care, we've got a real puzzle on our hands
895
Alex Armlovich
@aarmlovi
Check out my colleague
@lawsonhmansell
's work on this! Many states have repealed CON laws but some still have them. Then, beyond CON, there are real land use & permitting barriers for new or expanded facilities
247
SomeDude
@Long2007
why is that comparable? presumably you hire a traveling nurse on as temporary a basis as possible to fill some understaffed need that is important for your legal obligations/reputation/whatever. so i would imagine you do lose money on them, excluding those longer-term effects.
46
@Globaltom@mstdn.ca
@GlobalTom
Subtle point: a resident receives training, which means those training them have less time to spend doing other things. (What's quicker - doing rounds with a resident alongside you or without?)
154