A BALLOT QUESTION mandating nurse staffing levels in hospitals and other health care settings would increase health care costs in Massachusetts by $676 million to $949 million, according to an analysis released Wednesday by the state Health Policy Commission.

The report estimated that 2,286 to 3,101 new nurses would have to be hired under the law.

The report is a blow to the Massachusetts Nurses Association, which worked to put the question before voters, and is sure to further charge the debate over Question 1, which is already destined to be the most contentious – and most costly – measure on the November ballot.

The ballot question backed by the nurses’ union would require minimum nurse-to-patient ratios in various patient settings, with lower ratios in units treating more severe cases. The union argues that the measure would enhance patient safety, which it says is sacrificed by hospitals that don’t staff units with enough nurses to provide quality care.

The state’s hospitals are vehemently opposed to the measure, which they say would drive up health care costs without enhancing patient outcomes.

The Health Policy Commission report estimate of the ballot measure’s impact is in line with a report commissioned by the hospital industry, released in April, that said the measure would cost $1.3 billion in the first year and $900 million a year after that.

“This is an absolutely terrifying day for health care in our state,” said Dan Cence, a spokesman for the Coalition to Protect Patient Safety, the hospital-backed group opposing the question. “The Health Policy Commission’s analysis confirms that the negative consequences are too great and the costs are too high for rigid, government mandated nurse staffing ratios in the Commonwealth. This puts the cost question to bed.”

The Mass. Nurses Association released its own analysis last month, limited to acute care hospitals in the state, pegging the cost of the ballot question at less than $47 million. The analysis, conducted by Judith Shindul-Rothschild, a nursing economist at Boston College, projected that hospitals would need to add 539 to 1,617 new nurses.

Julie Pinkham, executive director of the Massachusetts Nurses Association, talks to reporters following Wednesday’s presentation at the Health Policy Commission.

Julie Pinkham, executive director of the nurses association, slammed the state analysis following its presentation at a meeting of the Health Policy Commission. She characterized a lot of the cost assumptions as “pure unadulterated pork.” She called the report’s projection of wage increases that would result from increased demand for nurses “a lovely dream world, but not reality,” saying there is no shortage of nursing graduates in Massachusetts seeking jobs here.

However, the state report said Massachusetts has a tighter labor market for nurses than all but nine other states.

The analysis, carried out by David Auerbach, the Health Policy Commission’s research director, and Joanne Spetz, a professor at the University of California, San Francisco, included a comparison of the ballot proposal with a law passed in 1999 in California, the only state with mandated nurse staffing ratios.

The report said the Bay State measure goes farther than the California law on most measures, requiring lower nurse-patient ratios, including no provision for waivers for rural hospitals, and mandating much faster implementation of the law.

On standard medical or surgical floors, for example, California requires one nurse for every five patients. The Massachusetts measure calls for one nurse for every four patients in such settings.

The report said Massachusetts, in 2016, already had higher hospital nurse staffing levels than California. It also found that Massachusetts hospitals performed better than California hospitals on five of six “nursing-sensitive quality measures,” such as surgical site infections following colon surgery or urinary tract infections associated with a catheter.

Spetz said studies on the effect of the California law on patients outcomes have been “a little bit of a wash,” with some showing benefits and others not, leading to the conclusion that there has been “no systematic improvement in patient outcomes” associated with the measure.

While the report concluded that California law led to a 4 percent increase, on average, in nurse wages, it estimated that the impact on nursing wages in Massachusetts could be greater – 4 to 6 percent — because of differences such as the smaller ratios the ballot question calls for and fines it allows for noncompliance with the mandates.

The analysis said a provision of the Massachusetts ballot question that requires minimum staffing ratios to be met “at all times” will further increase costs and staffing requirements. The report estimated mandating that minimum staffing levels be maintained during nurse breaks or other times nurses aren’t on duty during a shift boosted the estimated increased staffing requirement by an additional 10 to 20 percent.

The report said the mandate to hire more nurses, if the ballot question passes, would hit hardest on community hospitals with a high proportion of Medicaid and Medicare patients. The analysis estimated that nurse staffing levels would have to be increased by 21 to 30 percent at those facilities. Academic medical centers, by contrast, would only need to boost nurse staffing by 5 to 7 percent, according to the study.

David Cutler, a Harvard health care economist and member of the Health Policy Commission, expressed alarm at the projected impact on community hospitals with a large share of public payers, which are already struggling because of their inability to get the same price increases as other hospitals.

“I sort of shudder to think,” he said, about the impact of “imposing additional costs on those institutions when already their revenues are not increasing anywhere near as rapidly as other institutions.”

Pinkham, the nurses’ association director, said the fact that community hospitals would be most affected by the new staffing minimums underscores the fact that there are “have and have-not” hospitals in the state. “Is that appropriate, because you live in a particular town or city, that we’re all going to say that’s perfectly acceptable for you to have less nursing care, even though your needs are just as great as anybody else?” she asked.

The Health Policy Commission report did not include information on emergency department nurse staffing, which it said was not available in a form that allowed it to be analyzed, or outpatient departments. Because of that, the cost estimates “are likely to be conservative,” the report said.

The analysis did identify potential savings from the ballot measure, estimating $34 to $47 million in savings from reduced hospital stays and lower adverse effect rates that might be expected from increased nurse staffing.

The report said hospitals would look to a variety of options to absorb the higher costs the ballot measure would impose, if passed. Those include decreasing hospital margins or assets, reducing capital spending, closing unprofitable service units, and negotiating higher reimbursement prices from commercial insurers, costs that could be passed on to consumers in the form of increased premium charges.

The report estimated that the measure, if in place last year, would have increased health care expenditures in the state by 1.1 to 1.6 percent.

Recent polling shows the electorate evenly divided on the issue, with 44 percent supporting the question and 44 percent opposing it in a WBUR poll released late last month.

The two sides aren’t just arguing over the merits of the ballot question; they are also at odds over the Health Policy Commission wading into the issue, with backers of the ballot question charging last week that any review by the panel could be tainted by earlier meetings commission staff members held with a hospital trade group that opposes the ballot drive.

Pinkham criticized the analysis for relying largely on information provided by hospitals.

She said the Health Policy Commission appeared to approach this issue very differently than it did its analysis of the proposed merger of the Beth Israel Deaconess Medical Center system and Lahey Health, for which the commission “refused to accept the hospitals’ data and did their own analysis.”

“This is a state agency charged with oversight that chose to weigh in, and in instead of conducting a well-vetted and thoughtful analysis, they elected to use data provided by the single largest opponent to Question 1,” said Pinkham. “This is the fox guarding the hen house.”

The health commission will take up the issue again on October 16 and 17 as part of annual hearings it holds on cost trends in health care spending in the state.