Editorial Comment—The Next 30 Years of Stroke for Patients, Providers, Planners, and Politicians
Quibble with the models or puzzle over the formula, but heed the message of Figure 1. Stroke is more common in the elderly, and the proportion of elderly in the United States is rapidly growing. When questions involve comparisons, absolute numbers are shunned in favor of crude and especially adjusted rates. But here the question is what numbers we—as providers and planners of health care—should anticipate in the coming years. Figure 1 and its rising absolute numbers should alarm us. Are we prepared for this 98% projected increase over the next 30 years, from 139 000 patients dying from stroke in 2002 to 275 000 in 2033?
Considering mortality rather than morbidity underestimates the number, and many elderly are less frightened by death than by the prospect of long-term disability from stroke. Using Medicare hospital claims, the CDC estimated that during 2000 a total of 445 452 hospitalizations among Medicare enrollees were attributed to stroke.1 Only 8.7% of these 445 452 patients died during the hospitalization. Data on death following hospitalization were not reported. Additionally, focus on fatal and nonfatal stroke ignores the even more ubiquitous vascular injury to the brain—such as with small infarcts, small bleeds, and white matter changes—that can erode function in the elderly without ever leading to signs and symptoms recognized as stroke.
The solution rests in primary prevention in people who have not had a stroke and secondary prevention in patients who have. Paradoxically, developing effective treatments for stroke may only worsen the problem by enriching the population with survivors whose risk of recurrent fatal or nonfatal stroke is higher than that of the rest of the population. Competing risks complicate prevention as a solution. Successful prevention may simply shift the types of strokes that are disabling and killing people. For instance, antithrombotic therapy may reduce the occurrence of ischemic stroke but increase that of hemorrhagic stroke. Currently the leading causes of death are heart disease and cancer. Whether stroke overtakes these two in the next 30 years is a matter of how successfully these diseases can be prevented. Fewer people dying of heart disease and cancer would mean more people at risk of stroke. Consequently, the authors’ call for “concerted efforts to reduce stroke mortality rates” is a competition with other investigators trying to reduce the mortality rate of other common diseases of the elderly.
The types of investigations needed to identify modifiable risk factors for stroke and to implement programs to control these risk factors define the fields of epidemiology and health services research. Why have the investigators of this report focused on death, jury-rigging an administrative database to their purpose? Why not use a population-based national database with patient-specific information on the occurrence and outcome of all strokes? Such a study would have yielded more useful information than the current report. The line of researchers waiting to use such a database would be long indeed, if only such a database existed and were readily available. Instead, inherently compromised studies will likely remain the standard to address important questions of stroke prevention.
Funding is part of the problem. Stroke researchers compete for dollars distributed to advance knowledge in many fields. Even among the stroke researchers, those studying prevention compete with those performing basic research and concentrating on treatment of acute stroke. In addition, recent attempts to protect the privacy of individuals provide additional disincentives to perform epidemiology and health services research. As opposed to current trends in the United States, the federal government should be looking for ways to encourage investigators to perform such research and to facilitate, not impede, its conduct. Perhaps a better balance can be found between sharing patient-specific information for the public good and protecting the rights of the individual, especially when the studies are observational and low risk.
So the solution for these projected numbers will involve the concerted efforts of not only stroke researchers and health care planners but also patient-advocacy groups and politicians. More money alone will not be enough. The solution will also require a fundamental revision of how research is performed with a reduction of barriers and disincentives. To the extent that preventative strategies fail and these projected numbers are correct, we need to develop the capacity to care for the anticipated flood of patients with their stroke-associated morbidity and mortality.