📄 medicine

Specialist Physician Composition and Hospital Supply Expense Structure: A Four-Year Panel Study of General and Tertiary General Hospitals in the Republic of Korea

This four-year panel study of 285 Korean hospitals reveals that supply expense ratios vary significantly based on hospital type, ownership, region, and specialist physician composition, suggesting these differences reflect clinical functions and operational contexts rather than mere inefficiency.

Original authors: Do Won Kim, Vasuki Rajaguru, Tae Hyun Kim

Published 2026-06-25

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Do Won Kim, Vasuki Rajaguru, Tae Hyun Kim

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a hospital not just as a place where doctors heal people, but as a massive, complex kitchen. In this kitchen, the doctors are the chefs, the patients are the diners, and the supplies are the ingredients, pots, pans, and disposable gloves used to cook the meals.

This study is like a four-year investigation into 285 of these "hospitals" (kitchens) in South Korea. The researchers wanted to answer a simple question: How much of the hospital's total money is spent on "ingredients" (supplies) versus everything else (like paying the chefs and the rent)?

Here is what they found, broken down into everyday concepts:

1. The "Ingredient Bill" Varies Wildly

The researchers found that, on average, about 23% of a hospital's total spending goes toward supplies (drugs, bandages, surgical tools, even food for patients). But this number isn't the same for everyone. It's like how a bakery might spend 40% of its budget on flour and sugar, while a steakhouse might spend only 15% on meat because it spends so much on the expensive grill and the chef's time.

2. The "Menu" Matters Most (Specialist Doctors)

The biggest factor determining how much a hospital spends on supplies is what kind of doctors work there. The study looked at 18 different types of specialists.

The "Heavy Equipment" Chefs (Thoracic Surgeons): Hospitals with more chest surgeons had the highest supply bills. Think of these surgeons as the ones who need the most expensive, single-use gadgets (like high-tech staplers and specialized tools) for every operation. The more of these chefs you have, the higher your "ingredient bill" goes up.
The "Complex Care" Chefs (Internal Medicine): Hospitals with many internal medicine doctors also had high supply costs. These doctors manage patients with many different chronic illnesses. It's like a kitchen that has to prepare complex, multi-course meals for diners with very specific dietary needs, requiring a wide variety of ingredients and monitoring tools.
The "Time-Intensive" Chefs (Neurology): Interestingly, hospitals with more neurologists (brain and nerve doctors) actually had lower supply bills relative to their total costs. This doesn't mean they use fewer supplies; it means their costs are driven more by the time the doctors and nurses spend with patients and the use of expensive machines (like MRIs), rather than disposable "ingredients." The "bill" is heavy on labor and diagnostics, not on the physical supplies.

3. The Size and Location of the Kitchen

Big vs. Small Kitchens: The huge, specialized "Tertiary General Hospitals" (the big teaching hospitals) had much higher supply bills (34.5%) than the smaller "General Hospitals" (21%). This is because the big hospitals handle the most complex, high-tech cases that require expensive materials.
City vs. Country: Surprisingly, hospitals outside of the major cities (non-metropolitan areas) had higher supply costs than those in the big cities. The researchers suggest this is like a small, rural grocery store paying more for its produce than a massive city supermarket. The big city hospitals can buy in bulk and get better deals, while smaller, rural hospitals pay a premium for the same items.

4. Who Owns the Kitchen? (Public vs. Private)

This was a major finding. Private hospitals consistently had higher supply bills than public hospitals.

The General Hospital Split: This difference was huge for regular hospitals. Private general hospitals spent significantly more on supplies than public ones.
The Big Hospital Exception: However, for the massive tertiary hospitals, it didn't matter if they were public or private; their supply bills were similar. The researchers suggest that the big hospitals are so heavily regulated and specialized that ownership doesn't change how they buy supplies, but smaller private hospitals have more freedom to choose their suppliers and service lines, leading to higher costs.

The Big Takeaway

The authors warn us not to look at a high supply bill and immediately think, "This hospital is wasting money!"

Instead, think of it like this: If you see a restaurant with a very high food cost, it might be because they are serving a 10-course tasting menu with truffles and gold leaf (complex surgery), not because they are throwing food away.

The study concludes that supply expenses are a reflection of what the hospital actually does.

If you have many surgeons using high-tech tools, your supply bill goes up.
If you are a small hospital in a rural area, you might pay more for supplies because you can't buy in bulk.
If you are a private hospital, your business model might lead to different spending habits.

In short, the "ingredient bill" tells you about the hospital's menu, its location, and its ownership, not just how efficiently it runs. To truly understand if a hospital is efficient, you need to know what kind of "meals" (patients and procedures) they are serving.

Technical Summary: Specialist Physician Composition and Hospital Supply Expense Structure in the Republic of Korea

Problem Statement
Hospital supply expenses constitute a substantial yet underexamined component of healthcare spending, often ranking second only to payroll. In the Republic of Korea, while treatment materials are explicitly visible within the national fee schedule, the structural variations in supply expenses across different hospital types remain poorly understood. Existing literature has largely focused on physician supply, service utilization, or supply-chain management, with limited research linking specific specialist physician compositions to hospital-level financial outcomes. Furthermore, supply expense structures are influenced by a complex interplay of clinical service mixes, procedure intensities, physician preferences, and institutional purchasing capacities. This study addresses the gap in knowledge regarding how specialist physician density and institutional characteristics (hospital type, region, and ownership) are associated with supply expense ratios in Korean general and tertiary general hospitals.

Methodology
This retrospective panel study analyzed data from 2019 to 2022, covering 285 general and tertiary general hospitals in the Republic of Korea, yielding 1,119 hospital-year observations. The study linked two national data sources: the Hospital Accounting Information Disclosure System (for financial data) and the Health Insurance Review and Assessment Service hospital resource database (for structural and personnel data).

Outcome Variables: The primary outcome was the supply expense ratio, defined as annual supply expenses (drugs, medical supplies, and meal materials) divided by total medical expenses, expressed as a percentage. A secondary outcome was the natural logarithm of total annual supply expenses.
Explanatory Variables: The study examined 18 specialty-specific specialist densities (e.g., thoracic surgery, internal medicine, neurology), calculated as the number of specialists per 100 general beds. Additional covariates included hospital type (tertiary general vs. general), region (metropolitan vs. non-metropolitan), foundation type (public vs. private), inpatient revenue share, nurse staffing grade, and imaging equipment capacity.
Statistical Analysis: The authors employed pooled ordinary least squares (OLS) regression with standard errors clustered at the hospital level. Three model specifications were tested:
- Model 1: Baseline model with 18 specialty densities and year fixed effects.
- Model 2 (Preferred): Added hospital-level covariates.
- Model 3: Further adjusted for log medical revenue to examine changes after conditioning on revenue scale.
- Interaction models were also used to assess whether ownership-related differences varied by hospital type.

Key Results
The mean supply expense ratio across the sample was 23.1%. The analysis revealed significant associations between specialist composition, institutional characteristics, and supply expense structures:

Specialist Composition:
- Positive Associations: Higher densities of thoracic surgery specialists ( $\beta = 2.94$ percentage points) and internal medicine specialists ( $\beta = 1.20$ percentage points) were significantly associated with higher supply expense ratios.
- Negative Association: Higher neurology density was associated with a lower supply expense ratio ( $\beta = -1.88$ percentage points).
- Non-Significant/Variable: Orthopedic density did not show a significant association with the ratio, despite the known high cost of implants, suggesting that orthopedic costs are distributed across other cost categories (e.g., nursing, anesthesia) rather than just supplies.
Institutional Characteristics:
- Hospital Type: Tertiary general hospitals had significantly higher supply expense ratios than general hospitals ( $\beta = -4.99$ for general hospitals).
- Region: Non-metropolitan hospitals had higher ratios than metropolitan hospitals ( $\beta = 1.80$ ).
- Ownership: Private hospitals had substantially higher supply expense ratios than public hospitals ( $\beta = 6.22$ ).
Interaction Effects: The ownership gap was most pronounced among general hospitals. Private general hospitals had significantly higher supply expense ratios ( $\beta = 7.68$ ) compared to public general hospitals, whereas no significant difference was found between private and public tertiary general hospitals.

Key Contributions

Granular Linkage: The study provides empirical evidence linking specific specialist physician densities to hospital-level supply expense structures, moving beyond aggregate physician counts to specialty-specific impacts.
Standardized Accounting: By utilizing standardized Korean hospital accounting categories, the study offers a comparable measure of supply expenses across institutions, overcoming the limitations of locally defined purchasing data.
Contextual Nuance: The research demonstrates that supply expense ratios are not uniform indicators of inefficiency but reflect distinct clinical functions (e.g., supply-intensive thoracic surgery vs. labor-intensive neurology) and institutional operating conditions (e.g., purchasing power in non-metropolitan areas).
Ownership Dynamics: The study highlights that ownership-related cost differences are not uniform across all hospital tiers, being concentrated primarily in the general hospital sector.

Significance and Claims
The authors claim that hospital supply expense ratios should not be interpreted solely as measures of inefficiency or waste. Instead, these ratios serve as indicators of:

Clinical Complexity: Higher ratios in tertiary hospitals and those with thoracic surgery/internal medicine densities reflect the inherent supply intensity of complex care and multimorbidity management.
Institutional Context: Variations by region and ownership reflect differences in purchasing conditions, economies of scale, and organizational objectives.
Policy Implications: The findings suggest that benchmarking hospital financial performance or developing resource allocation strategies must account for specialty composition and institutional role. The authors conclude that further research utilizing detailed clinical, financial, and procurement data is necessary to fully elucidate the underlying mechanisms of these associations, cautioning that the current observational, hospital-level data cannot separate price, volume, case-mix, and quality mechanisms.