Gerard F. Anderson is director of the Johns Hopkins University Center for Hospital Finance and Management. He is a professor of health policy and management and international health at JHU.
Mary Smith is likely to get very expensive and inappropriate care. She is a 75-year-old woman with arthritis, mild dementia, asthma, a history of breast cancer, and mild diabetes. Mary's situation is relatively common. People with five or more chronic conditions represent 22 percent of all Medicare beneficiaries and 69 percent of all Medicare spending. They see an average of 13 physicians during the year and fill approximately 50 prescriptions. They are 99 times more likely to have a preventable hospitalization than someone without a chronic condition, and 98 percent of all hospital readmissions within 30 days occur in Medicare beneficiaries with five or more chronic conditions.
It is not only older people who have multiple chronic conditions. Her son, Harry Smith, is only 50 but already has high blood pressure and high cholesterol. Over 1 in 4 Americans of all ages have multiple chronic conditions and they are associated with over two thirds of all healthcare spending. These patients are also the ones most likely to be hospitalized, to have an adverse drug reaction, or to get conflicting medical advice from different providers.
What is the problem? Why is the care for people with multiple chronic conditions so expensive and fraught with so many adverse outcomes? The problem is that the current medical care system is not organized to treat them.
It begins with the lack of an evidence base to treat people with multiple chronic conditions. The gold standard used by the National Institutes of Health and the Food and Drug Administration to determine what is clinically efficacious is the randomized clinical trial. The problem is that older people and people with multiple chronic conditions are routinely excluded from the randomized clinical trials. While their exclusion may provide better evidence on whether the drug, device, or procedure works in a person with only that disease, it does little to show if the drug, device, or procedure works in a person with that disease and other chronic conditions.
The physician is routinely forced to make an educated guess whether the finding that he/she reads in a medical journal would apply to Mary or Harry since both of them would have been excluded from the randomized clinical trial. Twenty-five years ago, we routinely excluded women and minorities from most clinical trials—now we routinely include them. One possibility is to expand the inclusion criterion to include them. In addition, there are alternatives to the randomized clinical trials including pragmatic clinical trials, sophisticated statistical analyses, and comparative effectiveness research.
Most specialists are well trained to treat a specific chronic condition. The problem occurs when the person like Mary or Harry has multiple chronic conditions. Often it is the patients or their caregivers who must coordinate care. Generally our medical education system does not train clinicians to coordinate care. Also, most clinical training occurs in acute-care settings (inpatient hospitals and hospital outpatient departments), but the most effective care for people with multiple chronic conditions occurs in the doctor's office.
The payment system is mostly a fee-for-service system that pays for specific services. It does not recognize the ongoing needs of someone with multiple chronic conditions and the need for care coordination. Work is progressing in this area, but the payment system remains primarily a system designed to treat acute illnesses.
The delivery system responds to the financial incentives created by the payment system. There are models to provide integrated care, and some of them have been successful while others have been unable to both improve outcomes and lower costs.
Finally, when we attempt to measure good and bad care, the problem becomes exacerbated when the person has multiple chronic conditions. Sometimes one disease can interact with another and what is appropriate care for one disease interferes with the treatment of another. In this case, there is a problem defining the most appropriate care.
Fortunately, clinicians and policy makers are becoming aware of these issues and solutions are being proposed. A challenge is that it takes time. Research started today typically takes over 15 years to diffuse into medical practice and physicians graduating today will practice for the next 30 to 40 years. There are vested interests in retaining the current payment and delivery systems and it is difficult to change the quality metrics without an adequate evidence base. Nevertheless, progress is being made. Clinicians, the public, health insurers, and policy makers need to know where the money is being spent and where the outcomes need to improve in order for change to occur.