Primary care clinicians should not say "so what." The point is that different diagnostic and treatment strategies for common disorders may have to be tailored to each of the two causative elements, genetic and environmental. This dual approach is among the messages of a pair of articles in this issue [2, 3].

By "genetic elements," we mean those details about each person's genetic makeup that influence his or her structure and functioning under "normal" conditions and in the presence of disease. For each disorder, we expect that one or more genes determine the character and severity of the disorder. This means that, ultimately, if we know what form of each gene a person has, we could predict susceptibility to the disorder, its potential severity, and even treatment approaches.

For "environmental elements," we have to distinguish two broad types. One is the set referred to as "human" elements: those that reflect the environment created by human activities (for example, cultural, social, and behavioral factors). The second set refers to "background" elements: those that include purely physical influences, such as sunshine (actinic irradiation), and biological influences, such as infections and parasites.

Using these definitions, we can understand that common human disorders represent a spectrum defined by the relative importance of genetic and environmental causative factors. At one extreme are disorders with a predominantly environmental contribution (for example, electrocution by lightning). At the other extreme are disorders with a predominantly genetic contribution (for example, Huntington disease). In between, representing a mix of environmental and genetic factors, we can include most common disorders.

One such common disorder is peptic ulcer disease or, more precisely, the complex of gastritis, ulcer, and gastric cancer. This complex involves 1) a clearcut environmental contribution of infection with the bacterium Helicobacter pylori; 2) a genetic contribution of susceptibility to the infection; and 3) a genetic contribution to the consequences of the infection.

Although a large body of evidence exists to indicate a genetic contribution to peptic ulcer disease [4, 5], evidence from different geographic, cultural, and age settings also suggests a role for multiple causative factors. The lines of evidence supporting genetic contributions include family studies showing an increased familial aggregation (most for blood relatives, less for spouses); twin studies (higher concordance among monozygotic twins than among dizygotic twins); blood group studies (increased frequency of blood group O and nonsecretor and possibly other genetic markers); and an increased frequency of physiologic abnormalities among genetic relatives, especially increased serum levels of pepsinogen I [4, 6]. How do we interpret these long-standing genetic and epidemiologic data in light of a compelling role for H. pylori infection?

Although gastritis, peptic ulcer disease, and gastric carcinoma have several causes, at least one group of persons has one or more of these problems; for them, the cause ultimately is a partnership, a collusion between the person and the bacteria in that person's stomach [7]. But, why are the bacteria there? Does their presence reflect something about the person or even the person's genetic makeup? And, further, is there something about the bacteria's genetic makeup that contributes to the fact of, or the nature of, the "partnership?"

Malaty and coworkers [8] studied the familial clustering of H. pylori infections and showed that for infected persons with gastritis, 68% of their spouses also had the infection, whereas for normal control index cases, H. pylori infection was present in the spouse only 9% of the time. The offspring of infected index cases were more likely to be infected than were the offspring of uninfected index cases, and the offspring infectivity rate was the same whether the mother or father was the infected index case. The offspring data are especially important because the initial infection appears to occur predominantly in childhood [9]. Malaty and colleagues [8] concluded that for transmission of H. pylori infection, genetic factors are less important than living conditions. In support of this concept, Malaty and colleagues [10] also showed that the prevalence of H. pylori is "almost identical in Hispanics and blacks and significantly higher than in whites" and that there is an "inverse correlation between H. pylori infection and educational level." The overall conclusion was that for at least some groups of people, the frequency of H. pylori infection reflects socioeconomic status more than the genetic relatedness of the at-risk persons. Taken together, these data raise the question of whether genetic susceptibility has any role in H. pylori infection and its consequences.

Answering that question in this issue, Malaty and colleagues [2] evaluated the concordance of twin status and the presence of H. pylori infection. Their patient cohort was representative of the Swedish Twin Registry. The results, keying off the high concordance for monozygous twins reared apart, provide data that human genetic factors contribute substantially to determining who will be infected with H. pylori and who will ultimately be at risk for the spectrum of gastritis, peptic ulcer disease, and gastric cancer. Their data also confirm that shared and nonshared environmental factors are important contributors to the presence of these conditions.

The mechanism underlying this susceptibility to infection is not yet clearly elucidated. On the one hand, Boren and colleagues [11] have concluded that Lewis(b) blood group antigen mediates H. pylori attachment to human gastric mucosa. This study, focusing on a North American cohort, suggested that "the availability of H. pylori receptors might be reduced in individuals of blood group A and B phenotypes, as compared to blood group O individuals." However, in a previous Finnish study, Hook-Nikanne and colleagues [12] observed no relation among ABO blood group antigens, secretor status, and H. pylori antibody levels.

Nomura and colleagues, reporting in this issue [3], studied 150 Japanese-American men (living on the Hawaiian island of Oahu) with gastric ulcers and confirmed two aspects of H. pylori infection. First, they documented that gastric ulcers, as well as duodenal ulcers, are causally related to this infection. Second, they documented a correlation between the level of antibody response to the bacterium and the likelihood of the ulcer. These conclusions warrant attention because they extend the indications for aggressive treatment of this infection and they suggest that earlier treatment is preferred.

However, several important questions remain. Once a person is infected with H. pylori, what determines whether he or she will develop a duodenal ulcer, a gastric ulcer, gastric cancer, or none of these? One possible answer lies in the nature of the organism. Yoshimura and colleagues [13] focused on the genetic differences in the infecting bacteria found in different patients. They concluded that the presence or absence of human disease (for example, ulcer or no ulcer) may be attributed to "virulence factors" encoded by genomic DNA of the offending bacteria.

The other possibility is that the genetic makeup of the host determines not only whether infection takes place but also the consequences of the infection. Although the spouses were infected at a high frequency [8], they did not have a dramatically increased frequency of peptic ulcer disease [4]. In addition, data from the Japanese-American men described above indicate that an increased pepsinogen I serum level is more predictive of duodenal ulcer than is infection with H. pylori [14]. The critical question thus devolves to two choices: 1) Are the physiologic abnormalities seen among patients with ulcer and their relatives inherited and do they sum with H. pylori infection to predispose to ulcer?; or 2) Is there still one more step, with the inherited susceptibility being merely a predisposition to a deranged gastroduodenal function subsequent to H. pylori infection [15, 16]?

Data from multiple, diverse populations (from mainland North Americans, Japanese-Americans on Oahu, Finns, and Swedes) reinforce the conclusion that the complex of gastritis, peptic ulcer disease, and gastric carcinoma is causally related to infection with the bacterium H. pylori. But the data also indicate that genetic contributions cannot be ignored: It is the siblings and offspring who have the highest risk if a person has an H. pylori infection.

We have focused on multifactorial gastroenterologic disease, but other conditions exist where the same concerns are likely to be relevant. For example, many other infectious disorders are likely to have an epidemiologic profile that is consistent with a genetic determination of susceptibility, especially tuberculosis, malaria, and bacterial endocarditis. Beta-globin sickle cell mutations influence the likelihood or virulence of malaria, and the evidence for a tuberculosis susceptibility gene increases almost daily [17]. Now is the time to begin thinking about how we make the "new genetics" an integral, even mundane element of the "new medicine"