Observations

Let’s say you’re a nurse considering jobs at two hospitals in the same city.

Both are hiring. Both promise competitive pay, supportive leadership, and a commitment to excellence, which usually means someone in marketing owns a thesaurus. What you actually want to know is simpler: What happens when I show up for work?

Will the charge nurse have patients? Will someone cover my breaks? How many new nurses are on orientation? How often does the unit run short? Am I walking into a reasonably functional workplace, or am I about to become the structural support holding up a system that should have collapsed six months ago?

Those are fair questions. They’re also questions the public record cannot answer.

I know because I built a tool to check.

The California Nursing Observatory collects publicly available hospital data and puts it somewhere nurses can actually use it. Think of it as a weather report for the place you might spend your next twelve-hour shift. It can tell you whether certain conditions look worth investigating. It cannot tell you whether you’ll get lunch.

This distinction matters more than I expected.

Take Doctors Medical Center and Memorial Medical Center in Modesto, two hospitals serving the same community but operating under different ownership models and with very different histories. I worked at Doctors for fifteen years, so I already knew what no spreadsheet could tell me about the place. The question was what someone without those years of experience could learn from public records.

Quite a bit, actually.

The Observatory compares registered-nurse hours per patient day, contract nursing reliance, staffed beds, and licensed capacity. Doctors reported approximately 12.92 RN hours per patient day in 2024; Memorial reported approximately 17.44.

At first glance, that looks straightforward. Memorial has the bigger number, so Memorial must be better staffed.

Not necessarily.

Those figures describe nursing labor across an entire hospital. They don’t tell you whether one unit has excellent coverage while another survives on crossed fingers and caffeine. They don’t account for differences in patient acuity, service mix, or how the hours are distributed. And they definitely don’t tell you what your assignment looks like at 3 a.m. when two people call out and the emergency department keeps sending admissions.

Hospital-wide RN hours are useful, but they aren’t a photograph of the bedside.

Now look at the capacity figures in the hospital comparison tool. Doctors reports 342 staffed beds out of 461 licensed beds. Memorial reports 316 staffed beds out of 419 licensed beds.

Those numbers tell you each hospital reported fewer staffed beds than licensed beds. They don’t tell you why.

A licensed bed is part of a hospital’s authorized capacity. A staffed bed is one the hospital reports having personnel available to operate. The difference could reflect staffing constraints, service changes, operational decisions, or other factors the available records don’t explain.

It would be tempting to subtract one number from the other and announce how many beds each hospital “closed.” That would also be a good way to make a claim the evidence cannot support.

There’s another complication: those capacity figures come from 2025, while the nursing-hours figures come from 2024. They can help build a broader picture, but they aren’t measurements of the same hospital on the same day, or even within the same reporting year.

That’s how a clean-looking number becomes a dirty conclusion.

Contract nursing data adds another piece. A hospital relying heavily on agency or registry nurses may be managing vacancies, turnover, seasonal demand, or a temporary disruption. Public figures can show how much reported nursing work came from contract labor, but they usually cannot tell you why.

A higher percentage might reflect instability. It might reflect a deliberate staffing decision. It might reflect something else entirely. Without additional information, treating it as a verdict would be dishonest.

Even facility identities can be more complicated than they look. A hospital’s reporting record can include more than one campus, meaning a single set of numbers may describe multiple physical locations. Doctors’ licensed-bed figure, for example, includes both its acute-care campus and a behavioral-health campus. Those records cannot tell you which campus experienced particular staffing conditions.

Some financial filings have been audited; others remain in process. A hospital can belong to the Observatory’s reviewed pilot without its latest financial filing being complete, which is why those pending records need to remain visibly identified instead of quietly receiving the same treatment as audited ones.

None of this makes the data useless. It means the data needs adult supervision.

The deeper problem is that the information nurses most need is barely there.

Public records generally cannot tell you the actual nurse-to-patient assignments on a particular unit. They can’t tell you how many breaks were missed, how much mandatory overtime was worked, whether the charge nurse carried patients, or how often someone floated to a unit they weren’t prepared to cover.

They don’t reliably show first-year nurse turnover, orientation length, educator availability, or the number of experienced nurses responsible for training everyone else. They don’t tell you whether essential care was delayed because a nurse had too many competing demands.

The public can see parts of a hospital’s structure. The people considering working there still can’t see the pressure inside it.

That matters because staffing isn’t just a head count. A hospital can technically fill its positions while burning through the experience required to make those positions functional. It can replace seasoned nurses with new graduates, eliminate dedicated educators, assign preceptors full patient loads, and call the resulting arrangement a successful staffing strategy.

On paper, people are present. Whether the system can keep functioning that way is another question.

This is where the Observatory becomes useful precisely because it refuses to pretend it knows more than it does. The point isn’t to manufacture a safety score or crown one hospital better than another. The point is to give nurses better questions and enough context to recognize when the answers matter.

If a hospital’s reported RN hours are lower than those of comparable facilities, ask why. If contract labor increased, ask what changed. If staffed beds fall well below licensed capacity, ask how the hospital defines that difference and what it means for the units where you might work.

If management says orientation is excellent, ask how long it lasts, who provides it, and whether the preceptor carries a full assignment while teaching.

And if no one can give you a straight answer, that tells you something too.

The underlying numbers come from California’s hospital financial reporting system and California’s public health-data portal. Both make valuable information available. Neither gives a nurse a clear view of what happens during an actual shift.

After building the Observatory, I’m less impressed by how much hospital information is technically public than by how much of the important stuff remains effectively invisible.

Nurses are expected to decide where to work, what risks they’re accepting, and whether a hospital can support safe, sustainable practice. Yet the available information is scattered across agencies, reported on different timelines, and missing the measures that would reveal how the workplace actually functions.

So yes, you can build a staffing picture from public data.

Just don’t expect the staffing to be included.

Everybody knows a free rider: the person who arrives at the potluck without a dish, fills a plate, and leaves with leftovers. Annoying when the shared resource is potato salad. Considerably more serious when it is the nursing workforce.

I spent fifteen years in labor and delivery at Doctors Medical Center in Modesto, where we trained new nurses while caring for high-risk pregnancies, critically ill newborns, and a substantial Medi-Cal population. Our educators and preceptors turned beginners into nurses capable of handling emergencies no textbook can adequately prepare you for.

Then those nurses left, often enough for the arrangement to earn its own nickname: the training pipeline to Kaiser.

To work as a nurse at Kaiser, you needed two years of experience. Kaiser paid roughly 50 percent more, but you had to burn through those first two years somewhere else. That somewhere else was us.

We hired new nurses, trained them, and worked beside them through the hardest years of their careers. Once they had the experience Kaiser required, they could leave for better pay. Kaiser got experienced nurses without having to create them. We got another vacancy and another new nurse to train.

We also lost a nurse who might have trained the next one.

Kaiser’s own staff-nurse postings continue to list two years of recent clinical experience as a basic qualification. But Kaiser did not invent the arrangement. It was simply the clearest place to watch a broader pattern unfold: one employer absorbs the cost of developing a nurse, while another hires her once that investment begins paying off.

To be clear, nurses do not owe their careers to whichever hospital hired them first. Better pay, better benefits, or a less punishing workplace are perfectly reasonable reasons to leave. The problem is the system those decisions reveal.

A hospital that trains new nurses helps maintain a workforce that every other employer in the region can draw from. When those employers do not contribute proportionately to developing that workforce, the training hospital subsidizes the rest of the market.

The problem is familiar beyond Modesto. In evidence submitted to the UK Parliament, the Royal College of Midwives described healthcare employers avoiding training costs by hiring midwives trained elsewhere. It called them free riders and warned that if enough employers adopted the practice, the number of newly trained clinicians would decline.

Researchers studying international nurse migration have identified the same dynamic between countries. Whether a nurse crosses an ocean or drives across town, somebody invested in her development, and somebody else benefits.

That investment has measurable consequences. A study involving 524 new graduate nurses found that a nurse residency program was associated with a drop in first-year turnover from 36 percent to roughly 6 percent and a substantial reduction in contract labor use. Training requires time, money, experienced staff, and sustained attention. Hospitals that provide it are doing more than filling their own vacancies.

The burden begins before graduation. When I taught vocational nursing, finding clinical placements was a constant struggle. Kaiser in Modesto did not take students from my program. Today, Kaiser describes a paid work-study program for upper-level nursing students from selected partner universities. That represents an investment in some future nurses, but selective partnerships do little for vocational programs left searching for somewhere their students can learn.

California’s Board of Registered Nursing reported that 70 nursing programs were denied access to a clinical placement, unit, or shift in 2018–19. By 2020–21, that figure had climbed to 128. The state’s vocational nursing board likewise acknowledged that clinical placements were scarce even before the pandemic.

That is how a community ends up with people who want to become nurses, schools prepared to educate them, and hospitals complaining they cannot find enough staff.

A California nurse-employer survey conducted by UCSF found that hospitals particularly struggled to recruit experienced nurses in specialties such as labor and delivery, critical care, emergency medicine, and the operating room, even as new graduates were more available. Limited specialty placements and broader training bottlenecks contribute to the mismatch. So does an employment market that prefers buying experience to building it.

The financial landscape makes the imbalance harder to ignore. State hospital filings show that Doctors carries a substantially heavier Medi-Cal burden than Memorial, while Kaiser’s Modesto operation draws from a markedly different payer mix and participates in Medi-Cal through a different organizational structure. A hospital serving more publicly insured patients can therefore be the same hospital absorbing the cost of clinical placements, orientation, and specialty training while better-resourced systems recruit the nurses afterward.

For a while, healthcare can keep borrowing from Peter to pay Paul. The arrangement survives only while somebody still has enough capacity to take the hit.

Another preceptor takes a student. Another experienced nurse orients a new hire. Another hospital absorbs those difficult first years.

Every experienced nurse lost to burnout, retirement, a better-paying employer, or a department closure represents more than a missing name on a staffing grid. She may have been the person who could teach a new nurse to recognize maternal collapse, stabilize a newborn, or notice a patient deteriorating before anyone else does. Lose enough of those people, and the system begins consuming its own ability to recover.

California has started acknowledging the problem. Since 2025, state law has required hospitals and clinics to work in good faith with certain public nursing programs seeking clinical placements. That protection should extend to vocational nursing programs as well. Hospitals should also publicly report how many students they accommodate, how many new graduates they hire, and what they invest in precepting, orientation, and specialty training. Communities deserve to know who is actually helping maintain their nursing workforce.

Some hospitals create nurses. Others consume them. Many do both, but rarely to the same degree. Asking which role a hospital plays is not about punishing nurses for leaving or pretending every facility has identical resources. It is about understanding whether the institutions drawing from a shared workforce are contributing enough to keep that workforce alive.

The real danger arrives when the hospitals doing the creating no longer have the staff, time, or financial cushion to keep doing it for everyone else. By then, the problem is bigger than a nursing shortage: the system is running out of its ability to make more.