Nurse Strikes and Bargaining

Nursing labour disputes in the United States are settled by collective bargaining between a union and an individual employer or hospital system, so terms are local rather than national. What travels is the reference point: once a large system in a region settles, the number is known to every nurse interviewing there.

The subject of these disputes has shifted. Staffing now features at least as prominently as pay, and the framing has hardened. Unions increasingly bargain over staffing levels themselves rather than over extra money for tolerating short staffing.

The rule that shapes every hospital strike

Health care has its own provision in federal labour law, and it explains why nursing strikes are announced in advance rather than sprung.

Before striking or picketing a health care institution, a union must give at least ten days' written notice to the employer and to the Federal Mediation and Conciliation Service, stating the date and time the action will begin. The notice can be extended by written agreement of both parties. A stoppage without it loses the protection of the statute, which leaves the nurses who walk out liable to be discharged rather than reinstated.

Two consequences follow, and they run in opposite directions. Nurses get a fixed window in which the employer knows exactly what is coming, which is often when a settlement arrives. The employer gets the same ten days to book replacement cover, at whatever the market charges for a nurse who must be in place on a named morning. It is also why so many nursing stoppages are announced as one-day or two-day actions with a stated end: the notice fixes the start, and a defined length lets nurses limit lost pay while still imposing the cost of cover.

The early 2026 stoppages

The first quarter of 2026 produced the largest concentration of nursing labour action since the pandemic: about 46,000 health care workers off the job in six weeks, on two coasts, ending in settlements rather than exhaustion.

In New York City, from 12 January 2026, around 15,000 nurses across three hospital systems walked out. Most settled within days. One employer, NewYork-Presbyterian, held out for 41 days. The resulting contracts carry raises of more than 12 percent over three years.

At Kaiser Permanente, from 26 January to 24 February 2026, some 31,000 workers struck for four weeks across Kaiser's operations. The settlement was 21.5 percent over four years, together with patient safety provisions the unions had made central to the dispute.

What the raises amount to

Percentages over multi-year terms are routinely misread. A rise of 21.5 percent over four years is roughly 5 percent a year compounding, meaningfully above the market given that average registered nurse pay rose about 3.0 percent in the most recent year measured, but not a different order of magnitude. The New York settlements, at more than 12 percent over three years, land closer to the going rate. See registered nurse pay.

What a strike buys, on this evidence, is a few points above what the market would have delivered anyway, plus the non-wage terms, which is usually where the substantive fight was.

It is bought at a personal cost first. Forty-one days without pay is a serious financial event for a household, and any honest account holds both facts. The nurses who struck won more than they would have without striking, and they paid for it themselves before the employer did.

The clause that is new

The New York contracts contain something that did not previously appear in these agreements: limits on how artificial intelligence may be used in nursing work.

The concern is specific rather than general. It is not that software exists on the unit. It is that algorithms scoring patient acuity can in practice determine how many nurses a floor is given, without any identifiable person accountable for the judgement. If an algorithm says a unit needs four nurses and the nurses on it say six, the question of who decides was simply unaddressed in most contracts until now. See patient acuity and staffing systems for how those tools work and where they are gamed.

This is the first time it has been written in and is unlikely to be the last. Candidates who asked about ratios can be expected to ask about acuity systems too.

Why staffing dominates the bargaining table

Staffing sits underneath most of the other disputed items. It affects the likelihood of violence on a unit, because a short-staffed floor has nobody free to de-escalate early. See workplace violence in nursing. It drives burnout and therefore turnover, and turnover produces the vacancies that cause the short staffing. See nurse burnout and nurse turnover and retention.

It is also the item with the strongest evidence behind it, which is why unions bargain over the number rather than over compensation for the number. See nurse staffing and patient outcomes, which also covers the four states that have legislated minimum ratios instead of leaving them to bargaining.

Scheduling terms travel with staffing terms. Where a state does not restrict mandatory overtime, a contract is the only place a limit can come from. See nurse overtime.

The cost of a stoppage to an employer

Every day of a work stoppage is covered by agency staff at crisis rates, on top of the disruption and the reputational cost. Set against the price of settling, a long strike is rarely the cheaper course, and the 41-day example is the demonstration. Travel and agency cover already runs about $66,081 per nurse per year above employed staff in ordinary conditions. See travel nursing and nurse staffing agencies.

Settlements also reset local expectations regardless of who was party to them, which makes a neighbouring system's contract a recruiting fact for every employer in the area.

Sources

Find your next nursing job here

100% free for nurses. New jobs every day. Travel, per-diem, and permanent positions nationwide.

See something out of date? Sign in to suggest an edit.