ARIZONA Gov. Bruce Babbitt leans back into the sofa, casually rests one black loafer on the edge of the coffee table in his office, and speaks with the frankness of a man bearing up well under siege.
''We bit off too much too quick,'' he acknowledges rather ruefully.
The subject: the Arizona Health Care Cost Containment System (AHCCCS), designed to provide health care for some 170,000 indigent Arizonans.
With a growing concern across the country that the costs of medical care are, in Governor Babbitt's words, ''out of control nationwide,'' AHCCCS (known familiarly as ''access'') has become a closely watched experiment.
And just now the experiment is running rather roughly.
In a scant 18 months of operation, the program has seen a parade of hirings and firings of key personnel. It has mounted an astonishing $40 million cost overrun. Last month, the state finally took over complete control of program administration, which it had contracted out to a private administrative firm, McAuto Systems Group. McAuto, in turn, filed a suit against the state for $16.6 million in damages and unrecovered costs.
All of which has embroiled Governor Babbitt, a Democrat, in what Arizona Republic commentator Joel Nilsson calls ''his toughest political test'' - with even his longtime Democratic allies in the Republican-controlled legislature blasting away at his heels.
Why such political fallout? The immediate turmoil stems from the measures Governor Babbitt has proposed for what he calls ''clearing away the debris.'' In a bailout program outlined when the state took full control of AHCCCS administration March 16, the governor proposed to:
* Tighten eligibility standards for coverage by AHCCCS.
* Move to control the prices hospitals charge for treating indigents.
* Continue the current 1 cent sales-tax increase (due to expire June 30), and shift sales-tax revenues from counties to the state.
* Eliminate a $6 million teachers' merit pay plan, change the formula for teachers' retirement, delay the opening of several prisons, and raise the state's budget by $26 million.
His menu for changes, still moving through the legislature, raises some eyebrows. ''He'll make everyone unhappy,'' says William Meek, a former journalist specializing in health-care issues whose public-relations firm now represents McAuto. Among the disgruntled, he says, will be the counties (which stand to lose sales-tax revenues), the hospitals (which are lobbying hard against cost-capping), the legislature (which hardly wants to vote more dollars into AHCCCS), and the indigents themselves, who may find it harder to qualify for the program.
Beyond these local issues, however, are concerns of broader national significance. With America now spending some 10 to 12 percent of its gross national product for health care - more than it spends for defense - a viable cost-containment system is eagerly being sought.
The program here, modeled on the smaller-scale health maintenance organizations (HMOs) which have helped reduce medical costs around the country, seemed promising. Like an HMO, AHCCCS contracts for services at a prepaid rate based on head count. In contrast, the federally funded medicaid program provides medical help for the poor on a ''fee for service'' basis - with individual doctors, hospitals, and other ''providers'' billing the public for each case.
And where medicaid (available in every state but Arizona) is viewed as the child of a cumbersome, distant, and expensive federal bureaucracy, AHCCCS is home-grown and home-controlled - an important point in this conservative Western state.
Where, then, did it go astray? Observers here note that the experiment has several lessons to teach:
* The need for careful forethought. Arizona devoted only 10 months to planning before launching AHCCCS on Oct. 1, 1982. Having no precedents (since medicaid does not operate on a head count, or capitation payment system), the lawmakers had little to guide them.
* Concerns for accountability. Compared with medicaid, AHCCCS is difficult to monitor. Under medicaid, each case is on record - because the provider is not paid until the paper work is submitted case by case. Under AHCCCS, however, providers receive lump-sum contracts - which leaves state officials wondering whether the poor are actually receiving the health care to which they are entitled.
* The lack of sound data. Health-care specialists across the United States agree that the nation's health-care system is plagued by a lack of good numbers. ''There's very little real data about anything,'' says David Mundel, president of the Greater Boston Forum for Health-Care Action, a coalition of businessmen, insurers, and health-care providers working on cost-containment measures. He complains that policymakers have only ''snippets of data'' on which to base their judgments. That, agrees Governor Babbitt, is clearly a problem in Arizona: ''The data for the last year and a half,'' he says, ''is no good, in my opinion.''
Underlying these difficulties, however, is a fundamental question: Can the switch from a public-sector fee-for-service system (medicaid) to a private-sector bidding system (AHCCCS) really reduce costs? Describing the dilemma facing the Arizona legislature, an article in the fall 1983 issue of Health Affairs noted that ''the state must be committed to funding the system at the level of the winning bids'' - rather than trying to ''force these bids to fit, in total, within some predetermined budget figure.'' If the bids are too high, the article continues, ''there are only two options to pursue: reducing the comprehensiveness of future benefit packages, or tightening eligibility requirements.''
Arizona legislators, although uncomfortable with either option, are still confident that the experiment can be salvaged. The issue, Governor Babbitt concludes, is ''whether or not the HMO model can be imported into public health-care programs.'' So far, he says, the complexities of the experiment have taught him one thing clearly: that ''you can't run a medical services program the way you build a highway.''