That’s a big yes, according to a new research paper from the National Bureau of Economic Research (NBER).

The study shows that nurse strikes increase in-hospital mortality by 19.4 percent and 30-day readmission by 6.5 percent, numbers based on strikes in heavily organized New York State from 1984 through 2004. Patients who require extra attention are particularly affected and the study also shows that outcomes are just as negative for those patients admitted to hospitals that employ replacement or “scab” labor.

(For perspective: more than 15 percent of hospital employees are unionized; that’s six percent of American organized labor. And while unionization rates in most American labor markets have been in freefall for decades, union membership in hospitals is actually increasing, from 679,000 in 1990 to around 1 million in 2008. Outside of retail, construction, and manufacturing, healthcare workers go on strike more frequently than any other group of workers.)

The NBER paper’s findings are deeply troubling, if not particularly surprising. Nurses are an essential part of medical services. They are the people regularly closest to the patients, functioning as the eyes and ears of the a medical operation. A hospital without nurses simply won’t be able to comprehensively cover its patients and, given the nature of the work, it’s easy to see how a strike could be deadly. (Of the 38,228 patients admitted during strikes in the sample studied, 138 patients died who might have lived under normal conditions.)

If this study creates enough of a stir, we could see some states try to take steps to mitigate the dangerous consequences of a nursing strike. But should the labor rights of hospital staff be curtailed due to the nature of their work? If so, how should policy makers react? Three thoughts on this below the jump.

1. I could easily see some states simply outlawing strikes among hospital workers. There is plenty of precedent for such an action. In many states unionized police officers and firefighters are legally unable to go on strike because their duties are too important to risk prolonged absences.

2. Health care union laws could be reformed, although I’m not sure what fix could address the essential problem. Current law requires health care unions to provide a written notification to management and the Federal Mediation and Conciliation Service, pinpointing the exact date and time of striking or picketing, 10 days prior to the action. This provision gives hospitals time to prepare and find replacement workers. But the study shows that these temporary employees don’t have much affect on the in-hospital mortality or re-admission rates. I imagine that’s because the scabs simply don’t have the experience or the workplace relationships to work the job well. Extending the notification period wouldn’t change that.

3. Before significant reforms take place, there is another issue to consider. Labor lawyer/blogger Jeff Hirsch writes: “I also wonder what data is out there on patient care in unionized hospitals versus nonunionized ones. That's a slightly different topic, but has some relation to the policy questions involved.” True, and while a comprehensive study hasn't been conducted, to my knowledge, the NBER study’s authors note that unrelated researchers have found higher output rates and lower heart-attack mortality rates at unionized hospitals. I'm not surprised. Having a job with benefits, consistently decent pay, and worker representation usually results in less turnover, higher levels of productivity, improved workplace communication, and a better trained workforce. Policy makers should keep these benefits in mind. As the NBER study’s authors write, “Our results reveal a short-run adverse consequence of hospital strikes. These strikes may, however, contribute to long-run improvements in hospital productivity and quality…”