Toward the end of the 19th century, Abraham Jacobi, president of the New York Academy of Medicine, bemoaned the wave of "bacteriomania" that was then sweeping through the medical profession. European scientists and physicians had recently discovered bacteria and their role in human disease, and a frantic hunt was on for other bacterial agents of infection and disease. "The new theories that infectious . . . diseases have each their own bacillus, are so pleasant and promised to be so fruitful that it required some courage to critically resist the flood," wrote Jacobi. He could make the same complaint today.

Here at the beginning of the twenty-first century, a similar "microbiomania" is taking place. Driven by the availability of powerful molecular tools and the banishment by antibiotics and vaccines of many acute infections, medical researchers have turned their attention to the enduring mysteries of chronic diseases such as arthritis, cancer, and heart disease. Bacteria and viruses may cause many of these diseases, which is the theme of Paul Ewald's new book, Plague Time: How Stealth Infections Cause Cancers, Heart Disease, and Other Deadly Ailments.

It is exciting to think that some of our deadliest diseases could be beaten back with new antibiotics and vaccines. Ewald is a forceful proponent of that compelling notion. His new book, however, may be ahead of its time. Certainly it is ahead of the data.

Everyone's favorite example of an infectious agent causing a chronic ailment is the stomach-dwelling bacterium Helicobacter pylori, which causes stomach ulcers and gastric cancer. Australian physician Barry Marshall discovered it in the early 1980s. He gave himself an ulcer by drinking a flask of the bacteria to convince his colleagues that ulcers were really infections that could be treated with antibiotics.

Ewald uses the now famous Helicobacter story to explore possible links between other microorganisms and chronic diseases, and to bash the medical community for its diffident acceptance of the infection-chronic disease link. He cites other recent examples, including: hepatitis B virus (HBV) causing liver cancer, human papillomavirus (HPV) causing cervical cancer, and human herpes virus 8 (HHV-8) causing Kaposi's sarcoma. These are compelling examples and great medical success stories: antibiotics cure bleeding ulcers, the HBV vaccine prevents infection and subsequent liver cancer, and an HPV cervical cancer vaccine is now in clinical trials.

Leaving these solid examples, the author skates onto thin evidentiary ice when he tries to associate infections with complex diseases like arteriosclerosis, Alzheimer's, schizophrenia, and breast cancer. Research has uncovered some tantalizing associations between these diseases and various infectious agents, but association is not causation. Readers of Plague Time will do well to remember that, because Ewald often forgets.

Ewald makes a case for identifying a bacterium called Chlamydia pneumoniae as the likely agent of arteriosclerosis, a major coronary disease. The bacteria were found in the artery-clogging plaque of arteriosclerosis patients, and antibiotic treatment trials were quickly begun. The trials were disappointing. As recently as September, one researcher noted, "This infection theory, even if proven, is not going to give excuses for people to keep an unhealthy diet, or smoke." Further complicating the Chlamydia theory of heart disease is recent evidence that hepatitis A virus and cytomegalovirus (CMV) may also be involved in the disease.

Turning to the trauma of human breast cancer, the author fingers the ubiquitous Epstein-Barr virus (EBV) or maybe a mouse mammary tumor virus (MMTV) as likely causes. But again, ongoing research shows these links to be tenuous at best. Ninety percent of us are infected with EBV, and five studies have found no link between this virus and breast cancer. No one has managed to produce mammary tumors in mice by injecting them with human breast tumors, which should theoretically contain MMTV. DNA from the bovine leukemia virus has been detected in cancerous human breast tissue, but Ewald does not mention this virus and says little about the likely role of human endogenous retroviruses in cancers. Endogenous retroviruses are acquired through genetic inheritance, not infection, so perhaps they do not fit the book's paradigm.

The etiologies of mental illness and many neurological disorders are mysterious. Maybe they are infectious diseases. Ewald suggests Chlamydia may cause Alzheimer's and a horse virus called Borna may cause schizophrenia. Again, the data from current research are ambivalent.

Curiously, Ewald makes no mention of the well-documented role of prions (infectious proteins) in various neurological diseases, including their possible role in Alzheimer's, or the insidious toll one human prion variant is now taking among infected patients in the United Kingdom.

Studying causality in chronic diseases is a difficult and confusing business. Ewald has a solution. He writes, "We are left with scientific paralysis if we doggedly adhere to the standards of proof that have been established for acute infectious diseases. To converge on the most reasonable explanation, we will have to change the standards of proof that are required for acceptance of infectious causation of mental illnesses."

The design of clinical trials and the choice of technologies are going to change from disease to disease, but in the end, skeptics are still going to require a demonstrable, reproducible linking of cause-and-effect in order to design effective therapies and means for prevention for these diseases. Abandoning proof is not the answer to uncovering the etiologies of chronic illness.

Plague Time says little about the contribution of individual genetics, environmental pollutants, hormonal flux, lifestyle, or the aging of the immune system in initiating and sustaining chronic illnesses. Nor does the book say anything about the "hit and run" effects of some bacteria in causing chronic diseases such as rheumatic heart disease, post-infectious arthritis, and the neurological Guillain-Barré Syndrome. The vast number of common autoimmune diseases are also not discussed. Although some of these autoimmune diseases may be initiated by infections, it is hard to understand why the microbes would preferentially infect women who make up most of the autoimmune population.

Plague Time is an uneven book. The author jumps from one disease to the next without providing much supporting evidence for his infectious assertions, and there is a chapter on bioterrorism that seems wholly out of place in a book about the role of stealth infections in chronic diseases.

Toward the end of the book, Ewald returns to his earlier theme about the role of evolution in infectious diseases and low-tech approaches to controlling the transmission of highly virulent organisms. This was the theme of his insightful 1994 book, Evolution of Infectious Disease, for which he earned so much well-deserved praise.

Plague Time, however, is a collection of emerging ideas that is being pushed too hard and too fast. The evidence for infectious agents in many of the chronic illnesses Ewald describes is not yet firm. Even the well-documented Helicobacter-ulcer connection has been complicated by new research suggesting that while some strains cause ulcers and gastric carcinomas, eliminating other strains may increase the risk of esophageal cancer. The agents responsible for acute infections are easy to spot, but those responsible for chronic infections are just beginning to emerge. Plague Time is a book that should be written twenty years from now.