Hospital Star Ratings and Medicare Networks in Northeast Florida: How to Check the Hospital Before You Pick the Plan

“My Medicare Advantage plan listed my doctor, but now they say he’s out of network. How is that even allowed?” That question was posted publicly on a consumer Medicare Q&A site, and it is the single most common thing people call us about after 1 January. The version nobody posts is the one that costs more: the same thing happens with hospitals, and most people do not find out until they are in the back of an ambulance or sitting in a pre-admission office being told their plan sends surgical cases somewhere else.

Doctors are personal, so people notice when one disappears. Hospitals are infrastructure, so people assume they are all roughly the same and all roughly included. Neither assumption survives the published data. In the Jacksonville metro, the hospitals CMS rates on its five-star scale do not cluster. They spread.

Why does your plan quietly decide which hospital you end up in?

A Medicare Advantage plan is a contract between you, an insurer and a set of local providers. The insurer negotiates rates with hospital systems, and the systems it signs become the places you can use at in-network cost sharing. Use a hospital outside that set and you are either paying more or, depending on the plan type, paying everything. The network is not a courtesy list of nearby buildings. It is the boundary of your coverage.

That boundary is invisible while you are healthy, which is exactly when you choose the plan. You compare premiums, you compare dental and over-the-counter allowances, and then you sign. Eighteen months later a doctor says the words “we should get you admitted”, and the plan you picked on a kitchen table in October decides where.

There is a second layer people miss. Hospital systems own the physician groups, the outpatient imaging, the surgery centres and the rehab beds around them, so a plan that contracts with a system is usually contracting with that whole chain of care. When it does not, the gap is not one building. It is a corridor.

What a CMS hospital star rating actually is — and what it is not

CMS publishes an overall star rating for most Medicare-certified hospitals in the United States as part of Care Compare. It is a summary: CMS takes a set of measures the hospital reports, groups them into categories such as mortality, safety of care, readmission, patient experience and timely and effective care, weights those groups and reduces the result to a number from 1 to 5. The underlying measures are public too, and they are where the real information lives.

Now the part that keeps this article honest. A star rating is a facility-level average, and you are not an average. It says nothing about the surgeon who would do your procedure and nothing about the specific service line you need. It is built from data describing care delivered in earlier reporting periods, so it lags; a hospital that rebuilt a unit or fixed a readmission problem last year will not show it yet. And a 3-star hospital can be genuinely excellent at the one thing you are going in for, while a 5-star hospital may not perform that procedure at all.

We are also not going to tell you which hospital to choose, and you should be suspicious of any agency that does. What we can do is report what CMS published, date it, say how it was grouped, and hand you the method for checking it yourself.

The Jacksonville-area hospitals CMS rated in July 2026

A July 2026 pull of the CMS Care Compare hospital file returned 221 Medicare-certified hospitals in Florida — a complete state count rather than a capped page of results, which matters, because a truncated list quietly misrepresents a market. Eleven of those 221 facilities sit in Jacksonville-area ZIP codes. Here they are, as published.

Two caveats belong here rather than in a footnote, because a post about data quality that hides its own data quality problem is not worth publishing. First, the Care Compare tool’s city and county parameters were broken at the time of this pull: both returned errors, and the city field came back empty on every address. The grouping above was made by ZIP-code prefix from the returned addresses. Read it as “Jacksonville-area 322xx ZIPs”, not as a verified list of hospitals inside the City of Jacksonville.

Second, following directly from that: Ascension St Vincent’s St Johns County, in ZIP 32259, is in St. Johns County, not in the City of Jacksonville. It appears in the table because its ZIP begins 322, which is exactly the quiet error a ZIP-prefix grouping produces. We are leaving it in and labelling it, because deleting it would hide the method. Our St. Johns County Medicare review covers that market on its own terms.

Three of the eleven carry no rating. NH Jacksonville is a Department of Defense facility, and River Point Behavioral Health and Wekiva Springs are psychiatric hospitals. CMS does not publish an overall star rating for every facility type, so “not available” means the rating was not published, not that the facility scored badly.

How wide is the gap inside one metro?

Wide enough that the word “metro” stops doing useful work. Nine of the eleven facilities above carry a published rating, and they occupy four different points on a five-point scale.

Look at what that chart does and does not say. It does not say a 4-star hospital is 33% worse than a 5-star one, because star ratings are not a ratio scale, and it does not tell you where to have a hip replaced. What it does say is that the facilities a Jacksonville plan network might contract with are not interchangeable on the one measure CMS applies to all of them on the same basis.

Ownership shows up in the data, and it is still not a verdict

One pattern in the Jacksonville-area group is hard to miss. Eight of the nine rated facilities are voluntary non-profit, and the single proprietary acute-care hospital in the group carries the lowest rating in the group. That is what the file says, for that pull, in that ZIP grouping.

Here is what it does not say. Nine facilities is a small group, and one for-profit hospital in it is not a sample. Ownership structure does not determine the quality of a specific department, and there are strong proprietary hospitals and weak non-profit ones across the country. Note it, hold it loosely, and check the individual facility rather than the category.

Plan stars and hospital stars are two different programmes

This is where most readers get quietly misled, usually without anyone lying to them. CMS runs a star rating programme for Medicare Advantage and Part D plans, and a separate one for hospitals. They share a five-point scale and nothing else. Plan stars measure customer service, complaints, member experience surveys, drug safety and how often a plan’s members get recommended screenings. Hospital stars measure clinical outcomes, safety and patient experience at a building.

A high plan star rating tells you nothing about which hospitals that plan contracts with, and a low plan star rating does not mean its network is weak. In the CY2026 landscape file the mean overall plan star rating across the counties we serve runs from 4.19 in Duval to 4.26 in St. Johns, with Clay at 4.24, Nassau at 4.20 and Broward at 4.25. That is a tight band, and it is measuring plan administration, not the hospital you would be admitted to.

Put the two charts side by side and the point lands. Plan ratings vary by seven hundredths of a star across five counties; hospital ratings inside one of those counties vary by three whole stars. Compare only the first number and you are optimising the variable that barely moves. Our read on the Jacksonville plan market is in our Duval County Medicare review for 2026, and the same exercise for the Orange Park side of the river is in our Clay County review.

What does it cost when the network picks the hospital for you?

It depends on which side of the Medicare fork you are on, and the two sides fail differently. Under Original Medicare alone you can use any hospital in the country that accepts Medicare, which removes the network problem and replaces it with a cost problem: the Part A inpatient hospital deductible is $1,736 per benefit period for 2026, and Part B leaves you 20% of most covered services with no annual cap. A second admission later in the year can open a second benefit period and a second deductible. That is the gap a Medicare Supplement fills, and we compare the two routes in our Medicare Advantage versus Supplement guide.

Under Medicare Advantage the cost problem is capped and the network problem is real. Your in-network exposure stops at the plan’s maximum out-of-pocket, and for the 2026 plan year the median in-network out-of-pocket maximum across the First Coast counties is $6,750, against $3,900 in Broward County. Out of network, depending on plan type, that cap may not protect you at all. The standard Part B premium of $202.90 a month is owed either way, by everyone with Part B, regardless of which private plan sits on top.

What measured quality looks like when you go past the star

Hospitals are not the only thing with published quality data attached. CMS also publishes a HEDIS public use file for Medicare Advantage, reporting standardised clinical measures by contract number. For the 2025 data year, two Medicare Advantage contracts serving the Jacksonville area reported adults’ access to preventive and ambulatory health services of 94.89% and 96.43%, and colorectal cancer screening of 73.75% and 75.05%. Those are full-population measures on tens of thousands of members each.

Two honest notes about that table. The blood-pressure measures are hybrid measures drawn from a sample of 411 medical records against eligible populations in the tens of thousands, so the gap between 88.56% and 78.35% is a sample-based gap and should be quoted with its sample size or not at all. The antidepressant rows show something no marketing flyer mentions: both contracts manage roughly 85% of patients through the acute phase and only about two-thirds through continuation.

The file itself carries a limitation worth naming. It has no state, county, plan-name or organisation column, so it cannot be filtered to Florida directly. Getting a local read means pulling Florida contract numbers out of the landscape file first, then querying the quality file by contract. Any article claiming to show you “Florida Medicare Advantage quality data” without describing that step is showing you something it did not actually do.

None of this tells you which hospital your plan sends you to. It is the closest published proxy for whether a plan’s members get routine care at all, and it belongs in the same folder as the hospital list rather than instead of it. If you want a second set of eyes on the numbers, that’s what we’re here for.

The emergency department question almost nobody asks

A directory listing tells you a hospital is in network. It does not tell you the hospital has an emergency department. Those are different facts and the second one decides where an ambulance takes you. In the same July 2026 Care Compare pull, Broward Health Imperial Point in Fort Lauderdale is a rated acute-care hospital with no emergency department listed. A Fort Lauderdale retiree reading a plan directory would see the name, recognise it and reasonably assume it was a place to go in a crisis.

Two things follow. If you split the year between Northeast Florida and South Florida, you have two hospital lists to check rather than one. And wherever you live, the emergency-department column is a separate question you have to ask out loud, because no plan directory volunteers it. Ambulance routing is decided by emergency medical services on clinical grounds, not by your plan card, which is a good reason to know in advance which nearby facilities can receive you at all.

How to check the hospitals before you choose the plan

This is the part almost nobody does, and it takes about an hour the first time. Do the steps in order. The whole point is that the hospital list comes before the plan comparison rather than after it.

1. Write down the hospitals you would actually wantNot every hospital nearby. The one or two you would want for something serious, plus wherever your cardiologist, oncologist or orthopaedic surgeon admits. If you do not know where your specialist admits, call the practice and ask.
2. Look each one up on Medicare’s Care CompareGo to Medicare.gov and use Care Compare, the official tool. Search by name or ZIP code, open each facility, and note the overall star rating and the date the data covers. Do this before you have a favourite plan, so the number is not arguing with a decision you already made.
3. Read the measure groups underneath the starOpen the categories inside each facility page — mortality, safety of care, readmission, patient experience, timely and effective care. A hospital can carry a middling overall star and strong numbers in the group that matters for your condition.
4. Ask your own doctor the human version of the questionAsk which hospital they would send a family member to for your specific problem, and where they themselves have admitting privileges. A clinician’s answer and a published rating disagree often enough to be worth reconciling.
5. Ask each plan, in writing, about the coming plan yearCall the plan, name the hospital and the health system, and ask whether each is contracted for the coming plan year — not the current one. Ask for the answer by email. Directories go stale, networks reset on 1 January, and a verbal yes in November is worth nothing in February.
6. Ask the emergency-department question separatelyFor every facility on your list, confirm whether it has an emergency department. A directory listing does not tell you, and it is the difference between a hospital you can choose and a hospital you can be taken to.
7. Write the in-network out-of-pocket maximum next to each finalistFor the 2026 plan year the median across the First Coast counties is $6,750, and plans in this region range from $2,500 to $9,250. That column is your worst case, and it varies by thousands between plans that look identical on premium.
8. Only now compare the plansWith the hospital list settled and confirmed, compare what is left on drugs, doctors, out-of-pocket maximum and total expected cost in a heavy-care year. Then get a second opinion from someone with nothing to sell: call 1-800-MEDICARE, or Florida’s free counselling programme SHINE on 1-800-963-5337.

Bring your current plan and your prescription list; we’ll do the comparison with you. The person who arrives with a hospital list already confirmed in writing has done the part that takes the longest.

Before and after: one Jacksonville couple, on paper

Meet Ray and Denise, both 68, living off Beach Boulevard. They are invented, and the table below applies published 2026 figures to their invented situation. No carrier is named and no premium is quoted.

In the before column they compared two $0-premium plans on the dental allowance, picked one, and never looked at which hospital systems were contracted. Denise’s cardiologist admits to a system their plan does not contract with, which they discovered in March. In the after column they ran the eight steps above, got the network answer by email for the coming plan year, and chose a plan whose out-of-pocket maximum sits at the low end of the regional range.

What the after column did not require: paying more. Both columns are $0-premium plans. The difference came from sorting on different fields and asking two questions in writing before signing.

What a star rating still cannot tell you

Worth repeating at the end, because the middle of an article is where nuance goes to die. The overall star rating is a facility-wide summary built on measures from earlier reporting periods. It is not a statement about any individual physician, any particular procedure or any specific unit, and for many hospitals — including every psychiatric facility in the Jacksonville-area group — CMS publishes no overall rating at all.

Used carefully, it is still one of the few pieces of comparable, free, government-published information you have about where your plan can send you. Used carelessly, it becomes a number people wave at each other. The difference is whether you open what is underneath it. Things Medicare itself will not pay for are a separate list, and we keep that one in our post on what Medicare does not cover in 2026.

Where to get help that has nothing to sell you

Medicare.gov is the official government site, and its Care Compare tool is the source every figure in the hospital tables above came from. 1-800-MEDICARE is staffed around the clock. Florida’s State Health Insurance Assistance Program is SHINE, on 1-800-963-5337, run by the Florida Department of Elder Affairs; its counsellors are trained volunteers and no carrier pays them.

McDowell Business Resources is an independent agency, not an insurance carrier, and we are not affiliated with or endorsed by the U.S. government, the federal Medicare program or CMS. We do not offer every plan available in your area, and the information we give is limited to the plans we do offer. The method in this article works whether or not you ever call us.

How we help with the hospital question in Duval County

What we add is the labour and the paper trail. We take your list of doctors, your drug list and the hospital systems you name, and run all three against every plan we represent at your address. Then we do the part people dread: we get the network answers for the coming plan year in writing, so that when a directory changes you have something dated to point at.

We do it again every autumn, because contracts between insurers and hospital systems are renegotiated on their own schedule and nobody mails you a warning. The county detail lives on our Duval County page and the service outline on our Medicare page. There’s no cost and no pressure — book a free consultation and we’ll walk through it together.

What you get out of doing this properly

You stop treating hospitals as interchangeable infrastructure. You know, before you sign, which buildings your card actually opens and what CMS published about each of them. You know which nearby facility can take an ambulance. And you know your worst-case number before the year that produces it, instead of finding out in a pre-admission office.

The Annual Enrollment Period runs 15 October to 7 December for coverage starting 1 January, and if you are already in a Medicare Advantage plan, a second window from 1 January to 31 March allows one change. That is the calendar. The hospital list is the homework. Whatever you decide, decide it on the numbers. If you want help getting to them, we’re here.