BUILDING MORE hospitals does not necessarily mean better health care - or lower death rates. Sometimes, unfortunately, the dispersal of medical resources can mean just the opposite. Here is the Washington area, the campaign against unnecessary hospital construction is usually couched in terms of dollars and the burden on everybody's health insurance. But a recent note in the New England Journal of Medicine is a reminder that there are larger stakes here then mere money.

In the latter 1960s, Massachusetts' infant mortality rate was a good deal higher than it should have been in a wealthy state with superb medical resources. An inquiry by the state's medical society found that fully one third of the perinatal deaths - that of the fetus in late pregnancy or of the infant in the first month of life - were preventable. The study also found that preventable deaths were least common at those hospitals where the largest numbers of babies were born.

The state's health department responded by sharply raising mandatory standards for hospitals' obstertical services. Some of the smaller hospitals chose to stop delivering babies altogether. Did that mean worse care for the families living near those hospitals? There's no evidence of it. On the contrary, their babies are now being delivered in hospitals that meet rigorous rules of staffing and equipment.

Over the past several years, Massachusetts has gone further and set up a network of five centers for high-risk cases. These are in big metropolitan hospitals that command the full range of highly sophisticaeted (and expensive) resources that are needed only rarely - but, when they are needed, make the difference between life and death. When a doctor forsees trouble in delivery, he can move his patient to one of these centers. When unexpected trouble develops in a local hospital, the newborn infant can be rushed to the nearest center, in a specially equipped van, for care at a level as high as any in the world. In 1969, the state's perinatal death rate was 21 per thousand births. Five years later, it was down to 13.5 deaths. Other factors helped in this striking decline, and the number of infant deaths has been falling throughout the country. But the drop in Massachusetts was far faster than for the country as a whole, and the explanation is the improvement in regional organizaiton plus the elimination of underused services. In delivering babies, as in most other things, practice makes perfect.

The concept of regional perinatal centers was first worked out in places with vast rural stretches to serve - Quebec, for example, and Arizona. The experience of Massachusetts shows that the same principle works as effectively in highly urban areas. This case is only the latest among many demonstrating the enormous benefits in careful planning and regional cooperation that centralizes specialized care for high-risk cases.

Here in the Washington region, there have been repeated fights over proposals to build new community hospitals and scatter facilities more widely. Most of these projects are vigorously supported by their neighbors - people who feel that, in an emergency, a hospital would be a good thing to have nearby. But in the vast majority of medical emergencies, it isn't proximity to the hospital that counts. It's the range of resources and practiced skills that are waiting at the hospital when you get there.