Federal officials and insurance executives suspect the big 1967-77 shift from one health plan to another by federal employees will have a signifiance impact on the premiums they pay next year. But they don't know which way - up, down or stable - premiums will go.

Last October government workers and retirees in the world's largest "company" health insurance program had a short period in which to switch plans or options. For 1977 there are 60 plans in the government program. More then 220,000 federal workers and retirees have changed their plans, raising or lowering their coverage or trying to get better - or cheaper - coverage this year.

The big loser was Blue Cross-Blue Sheild. It is by far the largest, most popular of the plans available to the 3.2 million government workers and retirees in the federal health insurance program. That program, counting dependents, covers more than 7 million men, women and children. It has a major impact on hospital and doctor charges for everybody in the country.

The Blues' plan gained 12,216 new people. But it lost 85,101 who left it for less expensive plans, or for what individuals decided was better coverage. Most apparently went to the National Alliance of Postal and Federal Employees plan. That is available to all members of the predominantly black union, and to non member individuals who pay a small associate fee entitling them to enroll. The alliance plan jumped from just over 25,000 members in 1975, to more than 70,000 this year.

Aetna, the second-largest plan and (along with Blue Cross-Blue Shield) one of the so-called governmentwide plans available to everybody, had a net enrollment gain of just over 4,000 bringing its government membership to 484,296.

When it became apparent that Blue Cross-Blue Shield, because of premium increases, would lose members to improved benefit plans like the alliance, many people (including me) assumed that the Blues would lose the costly "high users" and the alliance would get them. It didn't work that way - making 1978 premium predictions even tougher.

Many assumed that the subscribers loss by Blue Cross-Blue Shield to the alliance and other union plans would help the Blues' program financially, and hurt the union and group health plans that "gained " people likely to be a drain on reserves.

But what has happened - and what nobody can figure out yet - is that most of the new members the alliance and other plans got were from the category of people who pay out more in premiums than they get back in benefits. They are either healthier, or younger, or more medically conservative. Anyhow, they use less medical services than "normal" insurance plan suscribers, or the "C" category of people who make heaviest use of health insurance.

What has happened, officials believe, is that Blue Cross-Blue Shield and Aetna, the big carriers, have retained the Type B (normal) users and Type C (heavy) users, while losing a good portion of their Type A (low) users to other plans.

What all this means is unclear at this point. It could mean even lower rates for some of the plans that picked up many Type A subscribers; and/or higher rates for those big plans that lost them and retained the people who make normal or heavy use of medical insurance.

It could mean that the smaller plans will hold the line on prices and benefits in 1978; or it could mean that the bigger plans will try to win back the Type A people by holding rates at current levels (or reducing them) and/or improving benefits. Or it could be that the big plans, if they suffer a financial drain, may have to raise rates for 1978.

Nobody yet knows what the changes mean. But the "experts" have been at least partly wrong in predicting health insurance trends. And that will certainly have an impact on premium rates most Americans - whether in government or private plans - pay next year.