Linking Public Health and Healthcare: What You Need to Know

By: Tiago Santana - Founder & CEO, Gray Group International • Serial entrepreneur and growth strategist who has built and scaled multiple companies across technology, media, and consulting. Expert in growth strategist and editorial voice for a global think tank building companies that advance the human experience

Key takeaways

• The short answer is cost pressure meets rising risk. Health outcomes are shaped far beyond clinic walls. The Robert Wood Johnson Foundation has long summarized evidence showing clinical care explains only part of health outcomes, while social and economic factors account for a large share. Many syntheses place those nonmedical drivers in the 30% to 55% range.
• Linking public health and healthcare matters because many costly drivers sit outside clinical care, yet both sides still pay for failure.
• Start with one problem that both sides can name in plain language. A strong use case has a clear population, one geography, one owner on each side, and one outcome that matters within six to twelve months. That keeps scope tight enough to test fast.
• Share the smallest set of data that helps someone act this week. That usually means geography, age band or risk group, service use trends, referral status, and a few social risk flags such as food access or transport barriers.

In 2024, Aisha Thompson led strategy at a nonprofit hospital in Brooklyn, New York. Her system had $2.1 billion in annual revenue and a 7.8% operating margin squeeze. Avoidable ER use was rising in three ZIP codes. Her team funded diabetes care inside clinics, while the city tracked food access and heat risk outside them. The data never.

In This Article:

• Key takeaways
• Why linking public health and healthcare matters now
• How do you start with one linked use case?
• Which data should be shared first?
• How do you fund the partnership?
• How do you know it is working?
• Ready to turn insight into action?
• Sources and further reading

Why linking public health and healthcare matters now

In short: In short: The short answer is cost pressure meets rising risk.

In short: The short answer is cost pressure meets rising risk. Health outcomes are shaped far beyond clinic walls. The Robert Wood Johnson Foundation has long summarized evidence showing clinical care explains only part of health outcomes, while social and economic factors account for a large share. Many syntheses place those nonmedical drivers in the 30% to 55% range.

Spending stays high even as needs rise. According to the OECD, average health spending across member countries was about 8.8% of GDP around 2019. The U.S. Spent far more, near 17% of GDP by CMS national health expenditure tracking. Public health still gets a small slice of total spend in many places, even though it often prevents the most expensive problems later.

A common mistake is treating these as separate missions owned by separate teams. That split wastes time and hides early signals. In our experience working with hospitals and city partners, the best programs start when everyone agrees on one measurable problem first.

Who feels the pain first?

We commonly see the same pattern across hospitals, payers, startups, and city agencies. One group owns claims data. Another owns surveillance data. A third owns trusted community relationships. Nobody owns the full outcome, so no one gets clear credit when things improve.

Aisha's team felt that gap clearly. Her hospital could show HbA1c levels and readmissions by patient panel. Local public health teams could map asthma burden, cooling access, and vaccination gaps by neighborhood. Community groups knew which buildings had broken elevators and which residents skipped visits because summer heat made travel unsafe. If you cannot explain who shares risk, who sees which data, and who acts first when trends worsen, you are not linking systems yet. You are coordinating by hope.

TL;DR: Linking public health and healthcare matters because many costly drivers sit outside clinical care, yet both sides still pay for failure.

How do you start with one linked use case?

In short: In short: Start with one problem that both sides can name in plain language.

In short: Start with one problem that both sides can name in plain language. A strong use case has a clear population, one geography, one owner on each side, and one outcome that matters within six to twelve months. That keeps scope tight enough to test fast.

In Brooklyn, Aisha's team picked uncontrolled diabetes in three ZIP codes because it showed up in ER use, pharmacy fills, food insecurity maps, and missed follow-up visits. That mattered more than a broad wellness program because it linked clinic work to neighborhood conditions right away. Our team typically recommends a simple filter: if you cannot describe the problem in one sentence without jargon, it is too big for phase one.

A common mistake is starting with ten metrics at once. Teams then spend months arguing about definitions instead of helping people faster. The better path is to define one shared workflow, one decision, and one review point. If the data cannot support action inside a quarter, it is probably not the right first use case.

What makes a good first use case?

Good use cases usually meet four tests: they are visible in both datasets; they affect money or capacity; they have an action within reach; and they have a clear comparison point after ninety days or less. For example, asthma hot spots near poor housing stock often work better than generic community wellness goals because housing teams can inspect units while clinics adjust care plans quickly.

Another useful test is whether both sides would notice if the problem improved tomorrow. If the answer is yes, the use case is likely strong enough to hold attention. If the answer is no, the project may sound important but still fail to drive action.

Which data should be shared first?

Put simply, In short: In short: Share the smallest set of data that helps someone act this week.

In short: Share the smallest set of data that helps someone act this week. That usually means geography, age band or risk group, service use trends, referral status, and a few social risk flags such as food access or transport barriers.

The CDC's National Syndromic Surveillance Program now receives de-identified emergency department data from more than 90% of U.S. Hospitals with emergency departments participating nationwide through ESSENCE-connected reporting streams in many jurisdictions (CDC reporting varies by state). That scale shows why speed matters more than perfect detail in early warning work. In our experience working with cross-sector teams, bigger datasets often slow trust down instead of speeding action up. People worry about privacy before they see value.

Keep the first share narrow enough that each partner can explain why it exists. If a field does not change a decision, leave it out for now. Smaller exchanges are easier to govern, easier to audit, and easier to expand later.

What should stay out of the first exchange?

Sensitive details that do not support the first intervention should stay out. That may include full clinical notes, unnecessary identifiers, and broad data dumps that create noise. Partners can always add more fields later if the workflow proves useful.

This is also where trust is built. When teams see that data sharing is limited, purposeful, and tied to a clear action, they are more willing to participate. Careful scoping is not a delay tactic. It is often the fastest path to adoption.

How do you fund the partnership?

In short: Funding is usually the point where good ideas slow down.

Funding is usually the point where good ideas slow down. Public health and healthcare teams often want the same result but sit in different budgets. Hospitals may save money from fewer admissions, while public health agencies may reduce future burden without seeing the direct savings. That means the business case must be practical, not theoretical.

A strong model links funding to avoided cost, reduced utilization, or lower pressure on scarce staff time. If one side pays and the other side captures the benefit, the agreement needs a clear reason to continue. Shared savings, targeted grants, and pilot budgets can all work if they match the size of the use case.

What financing signals matter most?

The most useful signals are simple: fewer ER visits, better follow-up rates, fewer duplicate outreach efforts, and lower cost per resolved case. These are the numbers that help leaders decide whether the work should scale. They also make it easier to talk across departments because they are tied to operations, not just mission language.

Do not wait for a perfect payment model before starting. Start with a small test and use the results to shape the next round of funding. That approach keeps the partnership grounded in evidence and reduces the chance that enthusiasm fades before the work proves itself.

How do you know it is working?

In short: A good evaluation plan tracks one outcome, one process measure, and one equity check.

A good evaluation plan tracks one outcome, one process measure, and one equity check. The outcome should match the original problem, such as fewer avoidable ER visits or better diabetes control. The process measure should show whether the workflow is being used. The equity check should show whether the benefit is reaching the intended group.

Teams often make the mistake of measuring everything and learning very little. A tighter scorecard is better. It tells leaders whether the shared data is changing decisions and whether those decisions are helping the right people.

Which metrics are worth reporting?

Report metrics that both sides can use in a monthly review. That may include referral completion, outreach success, missed visit rates, or neighborhood-level service gaps. If the metric cannot guide a next step, it is probably not worth reporting early.

Keep the review cadence short and regular. A monthly or quarterly check-in helps teams spot drift before the work loses momentum. It also gives partners a place to resolve problems while the project is still small enough to fix.

Ready to turn insight into action?

In short: Gray Group International works with [business](https://forbes.

Gray Group International works with business leaders to turn insight into action. Reading about the right approach is one thing; building the team, processes, and decisions that actually move metrics inside your specific organization is another. That second part is where most of the value lives, and it's where we focus.

Every engagement starts with a working session, not a deck. We listen to where you are today, look at the data and constraints with you, and propose the next two or three concrete moves that we believe will produce the most leverage. You leave with a plan you can act on whether or not you continue to work with us. Let's Connect

Sources and further reading

• United Nations - sustainability and global development