Andrea Niemi was motivated, no question about that. Bothered by weight problems most of her life, the 46-year-old St. Paul, Minn., lawyer two years ago had joined a health club that was about six blocks from her home. She had gotten herself into the kind of five-, sometimes six-days-a-week work-out routine that exists for most of us only as last year's unfulfilled New Year's resolution. But for all her discipline and sweat, the five-foot, five-inch Niemi had lost just 15 of her 210 pounds. She was still keeping storehouses of lovely dark chocolate in her desk at work and still buying those tasty malted-milk balls at the candy store on the way back from the courthouse. "I had gotten the health club exercise part of it together. I just couldn't get the eating angle," says Niemi.
Then, about a year ago, Marcia Hayes entered the picture. For the first few months, they started out chatting once a week for about 20 minutes. Niemi says she used Hayes "to bounce ideas off, to encourage me, to tell me when I was just being wacko." And instead of beating up on herself all the time and expecting perfection, Niemi says Hayes has encouraged her to take a more forgiving, Rome-wasn't-built-in-a-day approach to shedding weight. Eating a low-calorie diet and working out regularly, Niemi says she was hungry all the time. Hayes's advice? "She said maybe you should eat something!" The chocolate in the desk is gone and after-court candy store detours have stopped. Niemi now weighs 165. As a bonus, she and Hayes have come to know each other pretty well over the past year — well enough to laugh at little in-jokes, says Niemi: "She has gotten to know me and my issues and my progress."
But Andrea Niemi has not laid eyes on Marcia Hayes, and maybe never will. Hayes is a dietitian for Niemi's health plan, HealthPartners, and all of her patient counseling for HealthPartners is done over the phone. Far from resenting the facelessness of their encounters, Niemi says the convenience, the regularity and the one-on-one attention make it the best weight loss program she's ever been involved in — and Niemi feels as if she had tried them all.
And it didn't cost Niemi a penny beyond her standard premium. "I tell people who have HealthPartners that it is a great deal. And it's free!" says Niemi.
For more than 30 years now, the health and financial benefits of changing unhealthy behavior have been fairly obvious to physicians, health care providers, insurers and government officials. But the health care system's record on helping people like Niemi change their unhealthy ways is patchy at best, though some broad society-wide improvements are evident (see "Actually, We Are Living More Healthily," page 33). Mixed-up financial incentives, inattention to prevention by medical schools and a lack of motivation by patients have all gotten in the way. But with the evidence of the costs of changeable behavior so strong, the pressure on physicians, managed care organizations and the entire health care system to spend an ounce on prevention to save a pound of cure seems to be growing. The latest version of the Health Plan Employer Data and Information Set (HEDIS) measurements promulgated by the National Committee for Quality Assurance includes, for the first time, a measurement of how many smokers in a health plan were advised to quit by their physicians. The Health Care Financing Administration is considering a survey requirement that would, in essence, grade Medicare health plans on how well they keep members healthy.
"Physicians and health plans are going to be held increasingly accountable for outcomes," says Nancy Richardson, the vice president for business development for Healthtrac, a Menlo Park, Ca.-based company that sells self-care and health promotion counseling services. Richardson believes the trend will put some teeth into incentives to "do what is best for the patients and not just what costs less."
Truly embracing such an emphasis on behavior change would mean persuading physicians to view the practice of medicine in terms of populations instead of individual patients. Richard Thompson, M.D., director of preventive care at Group Health Cooperative of Puget Sound, says he wants the 680 doctors in the mixed-model HMO to start thinking of themselves as "practicing miniature public health." Through claims data analysis, physicians in health plans around the country are being told how their immunization and mammography rates stack up against their peers and against other standards. When it comes to influencing physicians, many health plans now seem to favor the the gentle written reminder. But they haven't forsworn the carrot or even the stick.
Besides influencing — even pressuring — their member physicians, managed care organizations are reacting to this heightened awareness about the connections between health and behavior by setting up telephone counseling services, offering self-care programs and literature and zeroing in on their members who are most at risk and most willing to change. To some extent, the old-style wellness program of giving everyone a discount on a health club membership or sending everyone nutrition advice is falling out of favor. "We have the found that the best thing to do is to target our investment and programs at the people who need and want the services, rather than sprinkling them across the whole population," says George Isham, M.D., medical director and chief health officer for HealthPartners in Minnesota. HealthPartners has beefed up its telephone counseling service so that it now handles 4,000 to 5,000 calls a month and employs the equivalent of 11 full-time dietitians and health educators.
Does this mean that managed care organizations will soon become behavior cops? Probably not, experts say. The evidence is still too blurry that interventions aimed at changing behaviors related to diet and exercise are effective. Health plans are moving cautiously, with smoking cessation programs leading the way.
"I think we mainly need to hold people responsible for their own behavior," says Robert Harmon, M.D., United HealthCare's national medical director. "But people need help. And a managed care organization needs to provide scientifically validated help so that people deal with their life style problems."
Whether through long-standing custom, bad habit, acquired addiction, simple carelessness or sometimes circumstances outside any one person's control, Americans routinely act with a kind of reckless disregard for their health. About one-third of adult Americans — 58 million people — are overweight. Six out of ten rarely or never exercise. One out of every four still lights up despite three decades of warnings that smoking causes lung cancer as well as a witch's brew of other diseases. Seat belt and child safety seat use is up, but by some estimates half of all Americans either ignore or don't correctly use car safety systems. And about 1.5 million are arrested each year for driving under the influence of alcohol. Others, obviously, elude arrest.
The American life style? Maybe it should be renamed the American disease-and-death style. Researchers have documented the havoc wrought by these behaviors in study after study. In 1993, William Foege, former head of the Centers for Disease Control and Prevention, and J. Michael McGinnis, head of the Office of Disease Prevention and Health Promotion in the U.S. Department of Health and Human Services, published a widely cited study in the Journal of American Medical Association about the "actual" causes of death (see chart at left). Instead of the usual groupings, such as heart disease and cancer, they used "external" (nongenetic) causes. Sifting through the medical literature, Foege and McGinnis calculated that roughly half of the 2.1 million deaths in the U.S. in 1990 were due to external causes. About four-fifths of those causes could be classified as behavioral: smoking, diet and activity level, alcohol, shootings, motor vehicle accidents and illicit use of drugs.
Perhaps to no one's surprise, tobacco use is deadliest of these behaviors. Foege and McGinnis figured tobacco use caused 400,000 American deaths, or 19 percent of the annual total. Meanwhile, the evidence mounts almost daily that inactivity and bad eating habits contribute to an enormous amount of illness and early death. The 1996 U.S. Preventive Services Task Force reported that physical activity and fitness reduce morbidity and mortality for six chronic conditions: coronary heart disease, hypertension, obesity, diabetes, osteoporosis and some mental disorders. Breast, prostate and colorectal cancer have all been linked to low activity levels. Just last month, Norwegian researchers published in The New England Journal of Medicine the results of a large prospective cohort study showing that women who exercise cut their breast cancer risk by 37 percent. Meanwhile, more and more connections are being made between specific dietary factors and cancer (see below).
The automobile is another American danger zone. Motor vehicle accidents are the leading cause of death among children and young adults, thus a leader in a statistic that many health policy makers believe is more revealing than simple death rates: years of life lost. Much of that highway carnage might be avoided if people wore lap and shoulder belts, which cut crash mortality rates in half. Drunken drivers are a public health disaster. In 1994, alcohol was involved in 41 percent of all traffic fatalities.
The dollar cost of all this unhealthy behavior can be sliced, diced, tallied and accounted for in dozens of ways. But the bottom line is always the same: Unhealthy behavior costs society a lot of money. Direct health care costs for smoking-related illness are over $50 billion per year. The annual tab for the country's drinking problems has been put at $70 billion (an estimate that includes lost productivity). The nation's cancer bill has also been estimated at $70 billion, and apparently much of that money could be saved if researchers are correct in their belief that two-thirds of cancer mortality is linked to modifiable behaviors such as smoking and diet. The National Safety Council estimated that in 1994 alone, motor vehicle accidents cost the country $169 billion, counting lost wages, medical expenses and administrative costs.
Put a different way, people who smoke and have lousy health habits end up spending more of the nation's medical dollar. The average lifetime medical cost in this country is an astounding $225,000 per person. Research has shown that medical costs for smokers, despite their shorter lives, are fully one-third higher than for nonsmokers. Health habits have been shown to be predictive of insurance claim costs. In one key study, the claims costs of people in the high-risk health habit group were eight times higher than people in the low-risk group.
Part of the problem with getting people to change their behavior is the problem with all delayed gratification: The effort is in the present tense, the benefit in the future. "It is hard to persuade a patient to do all of these things — and stop doing all of these things he enjoys doing now such as smoking, eating, being lazy — on the vague promise that maybe 10 to 20 years from now he won't have something happen to him. And it may not happen anyway," says H. Emerson Thomas, M.D., a Boston-based primary care physician who calls himself a "preventologist."
It is also a fallacy to picture all health-determining behavior as a matter of individual choice, argues Lauren A. Dame, a lawyer for the Washington, D.C.-based Public Citizen Health Research Group. In the United States and throughout the world, health and wealth march in virtual lock step: The richer you are, the better chance you have of a long and healthy life. Besides the obvious issue of more limited access to health care services, being poor brings with it a raft of conditions that are hostile to health: the neighborhood that is stressful, polluted and dangerous; the job that involves heavy lifting or working with dangerous chemicals; the store that sells junk food; the low self-esteem that feeds into drug abuse.
"There are life style choices that make a difference in health, but it is a smaller difference than we like to think," says Dame. "If poor people living near a toxic waste dump are unhealthy, it's not just because they don't go jogging every day."
Nor does society's effect on health-related behavior stop at the poverty line. Graham Colditz, a Harvard Medical School epidemiologist, has noted how contemporary American life steers people away from a simple, healthy activity that previous generations took for granted — walking. "The structure of our cities does not encourage any physical activity," he says. "While our grandparents walked to school and probably to work, we drive the car."
Even if patients were ready, able and willing to change their ways, the financial incentives can work against a health plan's investing much in prevention or behavior change. In many cases, says Thomas, "you have inexperienced, untrained physicians who don't have an incentive to do what they also have had little training to do," Thomas says.
For the managed care organization, prevention efforts pose what Helen Halpin Schauffler, an associate professor at the University of California-Berkeley School of Public Health, terms the "free rider" problem. A plan investing heavily in keeping people healthy may incur costs while never seeing benefits because the effects — avoiding serious disease and hospitalization — are not likely to show up until 10 or 20 years down the line, explains Schauffler. This is not true of all preventive services. Usually, childhood immunization is held out as an example of a preventive service with a quick payback in terms of avoided illness and medical cost. But particularly when it comes to modifiable behaviors that affect chronic disease, the benefits are so distal from the costs that the financial incentives are weakened.
Another difficulty is that prevention is not necessarily effective if the goal is reducing medical expenditures. Picking up a problem early through a screening test can result in years of fairly expensive treatment. In pure dollar terms, that cost will not necessarily be offset by the savings of avoided illness. A classic study in the mid-1970s showed how this worked for blood pressure screening. It is a cheap test, and medical treatment is often effective. But the savings generated by fewer blood pressure-related illnesses is not offset by the cost of years of medical treatment generated by the screening test.
"That is sort of a cynical notion — that if you don't get it back in dollars, why should anyone do it?" says Isham, at HealthPartners, noting that curative medicine is not held to that standard. Schauffler, pointing out that only 1 percent of the nation's health expenditures go to prevention, says, "Let's spend three percent and get healthier people. It's a no-brainer to me. But it is not a money-saver."
One of the impediments to efforts to change unhealthy behavior is lack of proof that physicians and other health care providers are very effective in this area.
So it is significant that the 1996 U.S. Preventive Services Task Force report echoes the original 1989 report's endorsement of smoking cessation programs. Given the fact that smoking is the most destructive changeable American health behavior, the unequivocal evidence that smoking cessation programs work is good news for anyone who wants doctors and managed care organizations to get more involved in keeping their patients healthy. The report recommends regular tobacco cessation counseling for all patients who use tobacco products. It also states that "a brief, unambiguous and informative statement on the need to stop using tobacco" is the most effective clinician message. Prescription of nicotine patches or gum is suggested as a way to ease withdrawal symptoms.
NCQA's new HEDIS measurement of how many physicians advised their smoking patients to quit obviously reflects the U.S. Preventive Services Task Force's recommendations and seems certain to spur managed care organizations to formulate better anti-smoking programs. Harmon, at United HealthCare, says that a program starting in July will include five telephone counseling sessions and discounts on nicotine patches and gum. But the jury is still out on how much of a difference smoking cessation programs can make. The most frequently cited study — a meta-analysis of 37 clinical trials — is now almost 10 years old. It found a modest 6 percent difference in quit rate between smokers in intervention programs and controls.
Nor is designing an effective program easy or cheap. Group Health Cooperative's "Free and Clear" program is often held up as a model of what a good smoking cessation program should be. Thompson has reported quit rates of about one in four. And rates of participation rose elevenfold when the smoking cessation program was made part of the HMO's standard benefits package in 1993.
The 1996 task force report also recommends that physicians counsel patients on dietary habits, activity levels and motor vehicle safety. But as the report made clear, these recommendations had more to do with the potential benefit than any solid evidence, based on clinical trials, that such counseling would work to change people's unhealthy ways.
Thompson is only slightly more sanguine. He says recent research has yielded some hopeful "glimmers" that physician counseling on nutrition might be effective. A study recently done at Group Health Cooperative showed that minimal nutritional counseling — basically, a physician handing out some literature — could result in small differences in dietary fat intake. But it was a small study with a very modest result. And when it comes to counseling patients on exercise, Thompson says, "The answers are just not there."
Researchers have been more successful in nailing down the effectiveness of physician intervention with problem drinkers, and in many ways task force recommendations on screening for drinkers parallel those made for counseling on tobacco use. In April, Wisconsin researchers published what they claimed was the first direct evidence that physician intervention with problem drinkers decreases alcohol use and utilization of health services. The intervention was pretty simple: two 10- to 15-minute counseling sessions by a physician using a scripted workbook. A year later, the 392 problem drinkers who had received the counseling were doing better on several problem drinking indices (binge drinking, recent excessive drinking, estimates of seven-day alcohol use). The counselees had also used the hospital less.
In theory, the spread of capitation should encourage behavior-focused preventive medicine. As Thomas points out, with capitation, "The message is that you don't get your rewards out of seeing sick patients and making them better. You get your rewards economically and professionally by keeping people well."
But capitation cuts both ways, Thomas adds. Counseling and educating patients are time-consuming, and capitation puts pressure on physicians to spend less, not more time, on each patient. "I used to schedule patient visits for 20 minutes," says Thomas. "We are now down to 15 minutes and we are looking at 10. And we are now considering reconfiguring our office so I no longer have a consulting room but exam cubicles instead, so I can pop in and move quickly."
While he admits that doctors can be a "little more slick in our operations," Thomas says there also comes a point where the "time provided is too little to accomplish the goal that is sought."
Capitation isn't the only way managed care organizations are prodding physicians to change. For the easy-to-measure, process-oriented preventive efforts, such as mammograms and immunizations, they are using claims data and other information to keep tabs on a doctor's performance, much as HEDIS does for the organization as a whole. Presumably, as smoking cessation and perhaps other behavioral change efforts become more a part of mainstream medical practice, physician performance in those areas will also be tallied. Thompson says Group Health Cooperative sends "clinical practice reports" to its staff doctors every three months that include information on immunization and mammography rates, as well as items more related to disease management. Harmon says United HealthCare sends "reminder letters" on immunization and mammography rates to physicians.
Thompson insists that the practice reports are less a report card than a communication in the spirit of "continuous quality improvement," though they do include benchmarks for comparison. Moreover, he says, Group Health Cooperative executives have discussed whether as much as 25 percent of a provider's salary should be contingent on meeting certain practice goals and patient satisfaction ratings. From fee-for-service to fee-for-time (i.e. salary), "what we are evolving toward is fee-for-outcome," says Thompson.
Having nurses or dietitians counsel patients on the telephone is cheaper than having physicians continually prod and cajole patients. The companies that are specializing in phone counseling and related services insist it can also be more effective. Ironically, the time pressure that managed care imposes on doctors may be part of the reason. "Physicians don't have the time, the tools or the training" to do such counseling well, says Richardson, the Healthtrac vice president.
Phone counseling programs have also increased their batting averages by being careful about whom they select, zeroing in on the people who are both most likely to change their health behaviors and at greatest risk of having serious health problems that could cost the health plan money.
In one project for a large New York City bank, Healthtrac used a familiar tool — the health risk questionnaire — and then combined the results with calculations about what conditions were going to be most expensive to treat in the next 12 months. HealthPartners' risk registry, which melds claims and patient survey data with biological information (weight, the results of lab tests), has much the same intent. Isham says physicians might use the risk registry to identify diabetics who need an exam or people for whom weight management might be a good idea. In many contexts, the distinction between health promotion through phone counseling programs and disease management does get fuzzy, especially when the "disease" is a condition such as obesity, which for many people is the consequence of behavior, namely poor diet and lack of activity.
In any event, the phone has a lot of fans in managed care these days. United HealthCare's smoking cessation program is conducted over the phone. Group Health Cooperative's "Free and Clear" smoking cessation program includes both phone and group counseling. Thompson has reported only a slight advantage — 35 percent compared to 28 percent — in the quit rate between the folks enrolled in the group sessions (who may be more committed to begin with) and those who choose to be counseled by a faceless voice over the phone.
Jackie Labat, manager of the HealthPartners' phone line services where dietitian Marcia Hayes works, says regular calls from a managed care phone center are especially helpful for people in weight loss programs because research has shown that continual contact is necessary to sustain weight loss. To help keep the counseling personal, Labat says, the plan member deals with the same dietitian or health educator all the time.
Andrea Niemi stresses how important it was to her success that she had an ongoing relationship with Hayes. "You can build an excellent rapport over the phone," notes Labat. Besides, says she, many members seem to prefer the convenience and privacy that the phone affords: "You are talking from the comfort of your own home." Or in Niemi's case, from the privacy of her office.
Prevention advocates have long argued that prevention should be measured by the benefit it pays in health, not the costs it might avoid downstream. But for those who can't afford to embrace that view, cost remains a key ingredient in evaluating programs aimed at changing behavior.
That's one reason why the group with the biggest ultimate emphasis on health behavior change may be neither health plans nor physicians. It may be the employers and government agencies who buy the policies and pay most of the premiums. In California, the Pacific Business Group on Health (PBGH), which includes some of the largest companies in the area, used its premium-paying clout to negotiate standards for a model benefit plan — a model that deliberately emphasized provision of preventive services.
The business group also put some financial bite behind its wishes: 2 percent of the premiums paid to health plans was made contingent on meeting certain performance standards (based for the most part on NCQA's HEDIS indicators). According to a recent account of the PBGH experience written by Schauffler and her colleague Tracy Rodriguez, in 1995, $7 million was at stake, approximately 30 percent of which hinged on meeting prevention performance measures.
In some instances, government health officials are pushing managed care organizations to get more involved in a keep-people-healthy approach. Immunization rates, because they are easy to measure and their payback in avoided medical care costs is patently obvious, are a favorite first target. In Arizona, a joint state-managed care effort boosted the infant immunization rate from 50 percent in 1993 to 77 percent today, according to Doug Hirano, special assistant to director of the Arizona Department of Health Services. Because all of Arizona's Medicaid population belongs to managed care plans, Hirano said, state officials knew a successful immunization effort would have to involve the plans delivering health care and not just depend on the traditional county health departments. The effort is based on an agreement by health plans to report their immunization rates to state officials. The resulting plan-to-plan comparison engenders "a little friendly competition," says Hirano.
But he adds that "Diet and physical activity are little bit more nebulous and more difficult to track. It is not clear how much inactivity and poor nutrition are costing a managed care plan. That is more direct with immunization or asthma. For public health to work cooperatively and collaboratively with managed care," concludes Hirano, "we both need to have incentives."