The state has rejected a proposal by Rhode Island Hospital to build a bone-marrow transplant unit to treat certain cancer patients, ending –– for now –– one of the longest and most contentious arguments ever to come before the Department of Health.

In the short term, the decision means that children who need bone-marrow transplants must still travel to Boston hospitals and that Rhode Island Hospital cannot move forward with a key element in its efforts to become a major academic medical center.

But the dispute over the proposal, which is sure to continue, goes to the heart of a decades-old conflict: the state’s many hospitals have competing economic interests, while policy-makers think patients would be better served if everyone could just get along. The decision suggests that state regulators are ready to push ever harder to make that happen.

Rhode Island Hospital’s proposal ran into trouble because there already is a bone-marrow transplant unit, for adults only, at Roger Williams Medical Center. That 14-year-old, five-bed unit has never operated at capacity, apparently because most people needing the treatment choose to get it from one of the prestigious hospitals in Boston.

Rhode Island Hospital argued that many of those people, rather than going to Boston, would prefer to undergo the therapy at Rhode Island Hospital, where it wanted to establish five beds for adults and two for children. Roger Williams officials called it a huge waste to build a new unit when an existing one lies half-empty just three miles away.

Under state law, Rhode Island Hospital had to prove that its proposed new service would be both “needed and affordable” –– that it would fill a gap in available services, and that it would be worth the cost, in this case $5.6 million to implement and about $10 million a year to operate. After 15 months of review, the Health Services Council, an advisory group to the health director, voted 7 to 5 on Tuesday to reject the proposal, and late Friday Health Director David R. Gifford upheld that decision.

George Vecchione, Rhode Island Hospital president, said yesterday that the hospital would re-file its proposal –– modified to ask for only two adult beds and two pediatric beds –– for the next round of reviews, which starts in June. He is also considering appealing Gifford’s rejection of the first proposal.

Bone-marrow transplantation treats certain forms of aggressive cancer, especially lymphoma and leukemia, and certain other blood diseases. The patient’s bone marrow is fully or partially destroyed, and replaced with new marrow cells, either donor cells or the patient’s own after processing.

Amid the review process, the Health Services Council instructed Rhode Island Hospital and Roger Williams Medical Center to come up with a way to collaborate on bone-marrow transplants.

They did. The result was a 13-point plan under which the hospitals would share a single stem-cell laboratory, located at Roger Williams, as well as sharing in treatment, administration and research. But they couldn’t agree on one point: Roger Williams wanted all the adult beds located at its hospital.

Rhode Island Hospital argued that it could not build the infrastructure to sustain a two-bed pediatric program without also having adult beds, and that sick children should have the option of staying close to home. But fewer than 10 children a year are expected to need the treatment.

Having the bone-marrow beds in-house is also considered essential before Rhode Island Hospital can win National Cancer Institute designation as a comprehensive cancer center, an honor that would draw top doctors and research dollars.

Roger Williams officials countered that everything Rhode Island seeks can be accomplished through collaboration with them, its smaller Providence neighbor.

When the final decision came before the Health Services Council on Tuesday, the debate lasted almost three hours. The council chairman, Robert J. Quigley, had voted against the proposal in a subcommittee meeting, but now proposed a modified version in which Rhode Island Hospital would get two adult beds instead of five; the pediatric beds could start right away, but the adult beds wouldn’t open for 18 months to give Roger Williams time to beef up its program. Additionally, the 13-point collaboration plan would go into effect.

Two other council members who had previously opposed the plan, including Larry Ross, the representative of Blue Cross & Blue Shield of Rhode Island, switched sides to support Quigley’s idea. Quigley’s motion failed, 5 to 7. Then the council voted by the same margin to reject the project.

Council member Wallace Gernt, who voted against the proposal, urged the two hospitals to continue to try working together. “I don’t think we can force you to collaborate, and we tried to do that,” he said.

Quigley lamented the negative vote. “We’re missing this opportunity for pediatric services and we’re missing a collaborative approach on this and other issues,” he said.

Gifford, the health director, said he studied all the evidence before deciding he agreed with the council vote. “Just because something is good for the applicant doesn’t necessarily mean it’s good for Rhode Island or meets need for Rhode Island,” he said.

Vecchione, the Rhode Island Hospital president, said he would file a letter of intent tomorrow indicating that the hospital intends to try again with its modified proposal. He said no decision had been made on whether the 13-point collaboration agreement would be part of the new application.

Kenneth H. Belcher, Roger Williams president, said he would continue to oppose having any bone-marrow treatment beds for adults at Rhode Island Hospital.

ffreyer@projo.com