Researching Another View of Healthcare Resource Utilization

While the population health pioneers discussed in my previous posts were exploring consumer-

driven facets of healthcare utilization (“Demand Management"), others researched the reasons

for the variation of healthcare resource consumption. As we shift toward value-based care,

tackling unwarranted variation is essential for improving outcomes and avoiding unnecessary

costs.

Wennberg’s Supply-Sensitive Care Insight

When we talk about value-based care, it’s easy to focus on payment models or technology

platforms. But decades ago, Dr. John Wennberg was asking a different set of questions: Why do

some communities perform so many more medical procedures than others? And what really

drives utilization?

In the 1970s, Wennberg and his colleagues at Dartmouth found that the supply of physicians and

hospitals in a region strongly influenced the amount of care delivered, specifically procedures

and hospitalizations. This became known as supply-sensitive care.

The key findings from Wennberg’s work (later codified in the Dartmouth Atlas of Health Care):

• More doctors → more procedures: Regions with more specialists and hospitals performed

more surgeries, diagnostic tests, and admissions, even when adjusted for population health status.

• Utilization varied widely: Rates of certain surgeries (like tonsillectomy,

hysterectomy, prostatectomy) varied up to 5- to 10-fold between similar

communities.

• Not driven by patient need: The differences weren’t explained by illness

burden, but rather by physician practice styles, local medical culture, and

resource availability.

• Implication for value-based care: This suggested that increasing supply alone

doesn’t necessarily improve health outcomes, it often just increases costs and

potential overuse.

In summary: The more physicians and hospital capacity a region has, the more procedures

patients receive and these differences weren’t explained by illness burden, but by supply, practice

styles, and local medical culture. This variability isn't just a U.S. phenomenon as the below

studies show.

Other Key Studies & Findings

1. Geographic variability in PET imaging for prostate cancer

• A recent Medicare-based study looked at PET use after prostate cancer

diagnosis across U.S. regions (metro, urban, rural). They found that PET

utilization rates by hospital referral region varied from ~2.2% to ~20.8%.

• Patients in metro counties underwent PET imaging more often (8.4%) than those

in urban (7.3%) or rural (7.2%) counties.

• After adjusting for other factors, the odds ratio for rural patients vs metro patients

getting a PET was 0.87 (95% CI: 0.82–0.92).

• The authors interpret this as evidence that capital-intense imaging (which

requires on-site scanners, radiotracers, specialized personnel) is less accessible

in lower-resource/rural settings.

2. Regional variation in diagnostic imaging (German study, Germany inpatient care)

• A study of German inpatient imaging practice looked at per-capita CT and PET

scans and sought to identify supply- and demand-side determinants of regional

variation.

• They found that supply-side factors (such as the density of imaging equipment,

hospital infrastructure) contributed to variation in imaging rates across regions.

• They conclude that availability of imaging capacity is a strong driver of utilization,

consistent with the broader supply-sensitive care model.

3. Physician workload & facility differences in PET scan reading

• From the IMV, a market research/industry intelligence report published by IMV

Medical Information Division, under the Science & Medicine Group:

• PET Market Summary Report: among sites with fixed PET/CT systems,

physicians read an average of 336 PET scans per year per site, but with large

variation:

• In hospitals with fewer than 200 beds: ~119 scans per physician

• In hospitals with >400 beds: ~567 scans per physician

• This suggests that larger hospitals, with more infrastructure, more staff,

more referral volume, tend to have much higher PET scan utilization per

physician than smaller ones.

• The variation in “scans per physician” based on facility scale is analogous to

variation in utilization seen in Wennberg’s work across regions with different

supply levels.

Why It Still Matters

Just as pioneers like Edington, Iverson, and Vickery redefined how we think about the reasons

patients seek healthcare resources, Wennberg’s work reminds us that achieving value isn’t about

“more” care, but about the right care at the right time.

• More doctors ≠ better health: Increased supply often leads to more procedures,

not necessarily better outcomes.

• Utilization is driven by availability: Imaging, surgery, and hospitalizations rise

where capacity exists.

• Value-based care must address variation: Tackling unwarranted differences is

key to aligning resources with patient health.

Question? How do you see supply-sensitive care shaping the next phase of value-based oncology and

population health strategies?

My next posts will look at the quality measures that were available in the late 1990s and early

2000s, how they evolved during the movement to value-based care, and the economic models

that were evolving at the same time.